feat(analyze): 新增 Ploss/Qfod 分析命令

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
Scottxjw
2026-08-14 15:36:29 +08:00
parent b88c07356e
commit 0667e2e9b8
8 changed files with 686 additions and 0 deletions

View File

@@ -12,6 +12,7 @@
| `aggregate` | 从本地库汇总生成发布库 | local.db → release.db | | `aggregate` | 从本地库汇总生成发布库 | local.db → release.db |
| `export` | 导出发布库到 CSV | release.db → CSV | | `export` | 导出发布库到 CSV | release.db → CSV |
| `clean` | 清理本地库数据(软删除) | 只写 local.db | | `clean` | 清理本地库数据(软删除) | 只写 local.db |
| `analyze` | 独立分析Ploss 余量/delta_p、Qfod 标定阈值 | log → 控制台报告 |
--- ---
@@ -71,6 +72,26 @@ dotnet run clean --all --confirm
--- ---
### 2.5 analyze
```bash
# Ploss 余量 / delta_p 分析(单文件,直接解析日志,不依赖本地库)
dotnet run analyze --type ploss --file <log>
# Qfod 标定阈值(两个文件:纯手机 / 手机+异物)
dotnet run analyze --type qfod --pure <纯手机.log> --foreign <手机+异物.log>
```
- `analyze` 为**独立分析**,不进入 `parse → aggregate` 发布管线,也不写任何文件(仅控制台报告)。
- Ploss 口径以实体语义为准Field9=ploss、Field10=threshold、Field12=FOD 标志):
- FOD 报警 = `Field12 == 1`
- 安全余量 `Margin = Threshold Ploss > 2000`
- 恒等式校验 `delta_p = Tx Rx pow_loss Ploss`(仅两行格式)
- delta_p 反解(两行格式且 FOD=1`delta_p' = DeltaP + 2001 + Ploss Threshold`,控制台输出重写后的 header 行
- Qfod 标定阈值:`Threshold = (纯手机 ΔQ 最大值 + 手机+异物 ΔQ 最小值) / 2`。因现无法自动区分两个日志文件,需显式传入 `--pure`/`--foreign`
---
## 3. 典型工作流 ## 3. 典型工作流
```bash ```bash

View File

@@ -0,0 +1,114 @@
# Ploss 分析delta_p 与安全余量判定)
> 本文档单独记录 Ploss 分析的分析口径与判定条件。
> 内容独立于既有需求文档 `docs/requirement/WCT数据需求分析.md`,不与原文档合并。
---
## 1. 核心口径
Ploss 两行日志格式:
```text
pow_loss = 2200, delta_p = 2500
FOD-> (16 个字段)
```
- **header 行**直接记录固件当前使用的 `pow_loss` / `delta_p`
- **FOD-> 行**记录 16 个字段,其中包含 Ploss功率损耗计算值、Threshold阈值、触发次数、FOD 结果等。
**权威字段语义**(以实体 `PlossRecord` 注释为准):
| 字段 | 语义 |
|------|------|
| Field9 | ploss - 功率损耗计算值 |
| Field10 | threshold - 阈值 |
| Field11 | 触发次数 |
| Field12 | FOD 结果标志 |
---
## 2. 判定条件
### 2.1 FOD 报警判定
**直接看 FOD 结果标志 == 1**。固件已完成判定Ploss>Threshold → 计数累积 → 满限置位),分析无需重建该逻辑。
### 2.2 delta_p 恒等式
```
delta_p = Tx Rx pow_loss Ploss
```
能量守恒恒等式,固件即用此公式计算 Ploss故必然满足可用于校验。
### 2.3 安全余量判定
```
Field10 Field9 = Threshold Ploss > 2000
```
正常充电时安全余量需大于 2000余量不足≤2000即异物或异常信号。
### 2.4 delta_p 调整反解(使余量 > 2000
当 FOD=1余量不足可依据 2.2 恒等式与 2.3 余量判定,反解出使余量严格 `> 2000`(整数即 `≥ 2001`)所需的 delta_p
```
margin = Threshold Ploss
delta_p' = DeltaP + (2001 margin)
= DeltaP + 2001 + Ploss Threshold
```
- `DeltaP`header 行记录的原始 delta_p
- `delta_p'`:调整后的新 delta_p
- 将 header 行重写为 `pow_loss = <原值>, delta_p = <delta_p'>` 后,按恒等式 Ploss 下降 `2001 margin`,余量达到 2001。
> **注意**:此反解仅在两行格式(有 header下有意义旧单行格式无 headerPowLoss/DeltaP 为 null无法调整 delta_p。
**测试数据字段位置**(与实体语义错位,以实体语义为准):
| 文件 | ploss | threshold | FOD 标志 |
|------|-------|-----------|----------|
| 实体语义(真实固件) | Field9 | Field10 | Field12 |
| ploss_test 实测 | Field8 | Field9 | Field11 |
| ploss_legacy 实测 | Field10 | Field11 | Field13 |
**ploss_test 中 14 条 FOD=1 行的计算结果**(新余量校验 = 2001
| 行号 | pow_loss | DeltaP | ploss | threshold | margin | delta_p' |
|------|----------|--------|-------|-----------|--------|----------|
| 247 | 2200 | 2500 | 798 | 1000 | 202 | 4299 |
| 259 | 2000 | 2200 | 760 | 750 | 10 | 4211 |
| 519 | 2200 | 2500 | 1308 | 1000 | 308 | 4809 |
| 527 | 3000 | 3500 | 1355 | 1250 | 105 | 5606 |
| 567 | 2200 | 2500 | 674 | 1000 | 326 | 4175 |
| 957 | 2000 | 2200 | 736 | 750 | 14 | 4187 |
| 1129 | 1000 | 500 | 123 | 350 | 227 | 2274 |
| 1265 | 2000 | 2200 | 497 | 750 | 253 | 3948 |
| 1533 | 3000 | 3500 | 1474 | 1250 | 224 | 5725 |
| 2157 | 2200 | 2500 | 361 | 1000 | 639 | 3862 |
| 2379 | 1000 | 500 | 397 | 350 | 47 | 2548 |
| 2413 | 3000 | 3500 | 955 | 1250 | 295 | 5206 |
| 2477 | 2000 | 2200 | 363 | 750 | 387 | 3814 |
| 2575 | 1000 | 500 | 560 | 350 | 210 | 2711 |
新 delta_p 范围 **2274 ~ 5725**(原 500 ~ 3500
> **局限**:示例日志为生成的测试数据,恒等式 `delta_p = Tx Rx pow_loss Ploss` 在数据中不成立(穷举全部字段组合均无法一致满足),故表中 delta_p' 为"若恒等式成立则应写入的目标值",无法用数据本身验证。真实固件数据(恒等式成立)下此公式精确成立。
---
## 3. 实测佐证
示例日志自动生成的测试数据按语义threshold ploss计算安全余量
| 格式 | 正常行 margin>2000 比例 | FOD 行 margin>2000 比例 |
|---|---|---|
| 两行 ploss_test | 917/1486 (61.7%) | 0/14 (0%) |
| 旧版 ploss_legacy | 361/593 (60.9%) | 0/7 (0%) |
FOD 行安全余量全部 ≤ 2000可区分正常/异物。
> 说明:示例日志为生成的测试数据,其字段位置与实体语义存在错位;**以实体语义为准**
> Field9=ploss、Field10=threshold

View File

@@ -0,0 +1,79 @@
# Qfod 标定分析Q值 → 标定 ΔQ 阈值)
> 本文档单独记录 Qfod 标定分析的分析口径、阶段区分限制与交付方式。
> 内容独立于既有需求文档 `docs/requirement/WCT数据需求分析.md`,不与原文档合并。
---
## 1. 核心口径
Q 值由日志直接记录,日志格式 `X#:Q A->B C D E F`
| 字段 | 名称 | 说明 |
|------|------|------|
| X | ChargerIndex | 充电器索引0, 1 |
| A | CoilIndex | 线圈索引0, 1, 2 |
| B | ΔQ (DeltaQ) | Q 值变化量 |
| C | CurrentQ | 当前 Q 值 |
| D | RawQ | 原始 Q 值 |
| E | FodType | 异物类型编码 |
**关键点**:「纯手机」与「手机+异物」分别对应**两个日志文件(两个测试场景)**
- 日志文件 A纯手机充电
- 日志文件 B手机 + 异物
标定分析需**先区分这两个日志文件**,再计算标定 ΔQ 阈值。
---
## 2. 标定 ΔQ 阈值计算
1. 分别从两个日志文件提取 ΔQ
- 文件 A纯手机取该文件 ΔQ **最大值**示例25
- 文件 B手机+异物):取该文件 ΔQ **最小值**示例40
2. 计算标定阈值:`Threshold = (25 + 40) / 2 = 32`
---
## 3. 分析步骤
1. 提取日志中 `#Q` 开头的行
2. 绘制 ΔQ 时序曲线,叠加 Y=Threshold 水平线
3. 验证余量:纯手机最高点 vs 阈值、有异物最低点 vs 阈值
---
## 4. 文件区分方式(现状与后续计划)
### 4.1 当前限制
暂无可自动区分「纯手机」/「手机+异物」两个日志文件的可靠依据,因此**当前版本无法自动计算标定 ΔQ 阈值**。
### 4.2 后续计划
日志文件将写清楚所对应的场景(文件名 Purpose 段或文件内标记),届时据此区分两个文件再计算标定阈值。
**待办**:定义日志文件区分标识,解析时识别并关联配对的两个场景。
---
## 5. 交付方式
- 标定 ΔQ 阈值属**独立分析****不进入发布库聚合**(发布库现有 Q值/Q基值 列仍为 CurrentQ/RawQ 全量平均,保持不变)。
- 计划通过**新建 CLI 分析命令**`analyze-qfod`)承载/输出,命令**尚未实现**。
---
## 6. 实测佐证
`data/test_input/singleMold-v1.0-hex2_1-iPhone15-qfod_test-20260710-1.log`(示例日志含 Phase 标记):
| Phase | 含义 | ΔQ 范围 | FOD Type(E) |
|---|---|---|---|
| 1 正常充电 | 纯手机 | 20~30均值 24.9 | 0 |
| 2 异物检测 | 手机+异物 | 35~55均值 44.6 | 1/2/3 |
| 3 边界值 | 临界区 | 30~35均值 32.5 | — |
| 4 负数 | 异常 | -25~0 | — |
> 说明:示例日志为生成的测试数据,仅作 ΔQ 数值区间的参考;真实数据中「纯手机」与「手机+异物」分属两个日志文件。

View File

@@ -0,0 +1,196 @@
using Gpulse.WCT.DataAnalyzer.Core.Application.Parsing;
using Gpulse.WCT.DataAnalyzer.Core.Domain.Local;
namespace Gpulse.WCT.DataAnalyzer.Core.Application.Analysis;
/// <summary>
/// Ploss 分析服务 - 安全余量判定与 delta_p 反解
/// 口径以实体语义为准Field9=ploss, Field10=threshold, Field12=FOD结果标志
/// 直接解析日志文件,不依赖本地库。
/// </summary>
public class PlossAnalysisService
{
private readonly PlossParser _plossParser;
public PlossAnalysisService(PlossParser plossParser)
{
_plossParser = plossParser;
}
/// <summary>
/// 分析单个 Ploss 日志文件
/// </summary>
public async Task<PlossAnalysisReport> AnalyzeFileAsync(string filePath)
{
var fileInfo = new FileInfo(filePath);
if (!fileInfo.Exists)
{
return new PlossAnalysisReport(
FileName: fileInfo.Name,
TotalCount: 0,
TwoLineCount: 0,
LegacyCount: 0,
FodCount: 0,
NormalAboveCount: 0,
NormalCount: 0,
FodAboveCount: 0,
IdentityHeldCount: 0,
IdentityCheckedCount: 0,
FodRows: [],
ErrorCount: 0,
ErrorMessage: "文件不存在"
);
}
var lines = await File.ReadAllLinesAsync(filePath);
var rows = new List<PlossRowReport>();
int errorCount = 0;
// 用于暂存两行格式的 header 行(与 ParseService 相同的合并逻辑)
string? pendingHeaderLine = null;
int pendingHeaderLineNumber = 0;
for (int i = 0; i < lines.Length; i++)
{
var line = lines[i].Trim();
if (string.IsNullOrEmpty(line)) continue;
var lineNumber = i + 1;
if (_plossParser.IsMultiLineStart(line))
{
pendingHeaderLine = line;
pendingHeaderLineNumber = lineNumber;
}
else if (_plossParser.CanParse(line))
{
if (pendingHeaderLine != null)
{
// 两行格式:合并解析
var result = _plossParser.ParseMultiLine(
new[] { pendingHeaderLine, line }, Guid.Empty);
if (result is { IsSuccess: true } && result.Record != null)
rows.Add(BuildRow(lineNumber, result.Record));
else
errorCount++;
pendingHeaderLine = null;
}
else
{
// 单行格式(旧格式兼容)
var result = _plossParser.Parse(line, Guid.Empty);
if (result.IsSuccess && result.Record != null)
rows.Add(BuildRow(lineNumber, result.Record));
else
errorCount++;
}
}
}
// 处理末尾遗留的 header 行(异常情况)
if (pendingHeaderLine != null)
{
errorCount++;
}
var fodRows = rows.Where(r => r.IsFod).ToList();
var normalRows = rows.Where(r => !r.IsFod).ToList();
return new PlossAnalysisReport(
FileName: fileInfo.Name,
TotalCount: rows.Count,
TwoLineCount: rows.Count(r => r.IsTwoLine),
LegacyCount: rows.Count(r => !r.IsTwoLine),
FodCount: fodRows.Count,
NormalAboveCount: normalRows.Count(r => r.Margin > 2000),
NormalCount: normalRows.Count,
FodAboveCount: fodRows.Count(r => r.Margin > 2000),
IdentityHeldCount: rows.Count(r => r.IsTwoLine && r.IdentityHeld),
IdentityCheckedCount: rows.Count(r => r.IsTwoLine),
FodRows: fodRows,
ErrorCount: errorCount,
ErrorMessage: null
);
}
private static PlossRowReport BuildRow(int lineNumber, PlossRecord record)
{
var isTwoLine = record.PowLoss.HasValue && record.DeltaP.HasValue;
var ploss = record.Field9;
var threshold = record.Field10;
var margin = threshold - ploss;
var isFod = record.Field12 == 1;
// 恒等式校验(仅两行格式有 header 数据):
// delta_p = Tx - Rx - pow_loss - Ploss
bool identityHeld = isTwoLine
&& record.DeltaP == record.Field5 - record.Field4 - record.PowLoss - ploss;
// delta_p 反解(仅两行格式且 FOD=1:
// delta_p' = DeltaP + 2001 + Ploss - Threshold
int? deltaPPrime = null;
string? rewrittenHeader = null;
if (isTwoLine && isFod)
{
deltaPPrime = record.DeltaP + 2001 + ploss - threshold;
rewrittenHeader = $"pow_loss = {record.PowLoss}, delta_p = {deltaPPrime}";
}
return new PlossRowReport(
LineNumber: lineNumber,
IsTwoLine: isTwoLine,
PowLoss: record.PowLoss,
DeltaP: record.DeltaP,
Ploss: ploss,
Threshold: threshold,
Margin: margin,
IsFod: isFod,
IdentityHeld: identityHeld,
DeltaPPrime: deltaPPrime,
RewrittenHeader: rewrittenHeader
);
}
}
/// <summary>
/// Ploss 单行分析结果
/// </summary>
public record PlossRowReport(
int LineNumber,
bool IsTwoLine,
int? PowLoss,
int? DeltaP,
int Ploss,
int Threshold,
int Margin,
bool IsFod,
bool IdentityHeld,
int? DeltaPPrime,
string? RewrittenHeader
);
/// <summary>
/// Ploss 分析报告
/// </summary>
public record PlossAnalysisReport(
string FileName,
int TotalCount,
int TwoLineCount,
int LegacyCount,
int FodCount,
int NormalAboveCount,
int NormalCount,
int FodAboveCount,
int IdentityHeldCount,
int IdentityCheckedCount,
IReadOnlyList<PlossRowReport> FodRows,
int ErrorCount,
string? ErrorMessage
)
{
public bool IsSuccess => ErrorMessage == null;
}

View File

@@ -0,0 +1,123 @@
using Gpulse.WCT.DataAnalyzer.Core.Application.Parsing;
namespace Gpulse.WCT.DataAnalyzer.Core.Application.Analysis;
/// <summary>
/// Qfod 标定分析服务 - 由「纯手机」与「手机+异物」两个日志文件计算标定 ΔQ 阈值
/// Threshold = (纯手机 ΔQ 最大值 + 手机+异物 ΔQ 最小值) / 2
/// 直接解析日志文件,不依赖本地库,也不进入发布库聚合。
/// </summary>
public class QfodCalibrationService
{
private readonly QfodParser _qfodParser;
public QfodCalibrationService(QfodParser qfodParser)
{
_qfodParser = qfodParser;
}
/// <summary>
/// 计算标定 ΔQ 阈值
/// </summary>
/// <param name="pureFile">纯手机充电日志文件</param>
/// <param name="foreignFile">手机+异物充电日志文件</param>
public async Task<QfodCalibrationReport> CalibrateAsync(string pureFile, string foreignFile)
{
var pureSummary = await ReadSummaryAsync(pureFile);
var foreignSummary = await ReadSummaryAsync(foreignFile);
if (pureSummary.ErrorMessage != null)
return new QfodCalibrationReport(null, null, pureSummary, null, 0, 0, 0, pureSummary.ErrorMessage);
if (foreignSummary.ErrorMessage != null)
return new QfodCalibrationReport(null, null, null, foreignSummary, 0, 0, 0, foreignSummary.ErrorMessage);
if (pureSummary.Count == 0 || foreignSummary.Count == 0)
{
return new QfodCalibrationReport(
pureSummary.FileName, foreignSummary.FileName,
pureSummary, foreignSummary, 0, 0, 0,
"日志文件中未提取到 Qfod 数据(无 #:Q 行)");
}
var threshold = (pureSummary.Max + foreignSummary.Min) / 2;
return new QfodCalibrationReport(
PureFileName: pureSummary.FileName,
ForeignFileName: foreignSummary.FileName,
Pure: pureSummary,
Foreign: foreignSummary,
Threshold: threshold,
PureMargin: pureSummary.Max - threshold,
ForeignMargin: foreignSummary.Min - threshold,
ErrorMessage: null
);
}
private async Task<QfodFileSummary> ReadSummaryAsync(string filePath)
{
var fileInfo = new FileInfo(filePath);
if (!fileInfo.Exists)
{
return new QfodFileSummary(fileInfo.Name, 0, 0, 0, 0, $"文件不存在: {filePath}");
}
var deltaQs = new List<int>();
var lines = await File.ReadAllLinesAsync(filePath);
foreach (var line in lines)
{
var trimmed = line.Trim();
if (string.IsNullOrEmpty(trimmed)) continue;
if (!_qfodParser.CanParse(trimmed)) continue;
var result = _qfodParser.Parse(trimmed, Guid.Empty);
if (result.IsSuccess && result.Record != null)
deltaQs.Add(result.Record.DeltaQ);
}
if (deltaQs.Count == 0)
{
return new QfodFileSummary(fileInfo.Name, 0, 0, 0, 0, null);
}
return new QfodFileSummary(
FileName: fileInfo.Name,
Count: deltaQs.Count,
Min: deltaQs.Min(),
Max: deltaQs.Max(),
Average: deltaQs.Average(),
ErrorMessage: null
);
}
}
/// <summary>
/// 单个 Qfod 文件的 ΔQ 统计
/// </summary>
public record QfodFileSummary(
string FileName,
int Count,
int Min,
int Max,
double Average,
string? ErrorMessage
);
/// <summary>
/// Qfod 标定报告
/// </summary>
public record QfodCalibrationReport(
string? PureFileName,
string? ForeignFileName,
QfodFileSummary? Pure,
QfodFileSummary? Foreign,
int Threshold,
int PureMargin,
int ForeignMargin,
string? ErrorMessage
)
{
public bool IsSuccess => ErrorMessage == null;
}

View File

@@ -2,6 +2,7 @@ using Microsoft.Extensions.Configuration;
using Microsoft.Extensions.DependencyInjection; using Microsoft.Extensions.DependencyInjection;
using Gpulse.WCT.DataAnalyzer.Core.Infrastructure.LocalData; using Gpulse.WCT.DataAnalyzer.Core.Infrastructure.LocalData;
using Gpulse.WCT.DataAnalyzer.Core.Application.Parsing; using Gpulse.WCT.DataAnalyzer.Core.Application.Parsing;
using Gpulse.WCT.DataAnalyzer.Core.Application.Analysis;
using Gpulse.WCT.DataAnalyzer.Core.Application; using Gpulse.WCT.DataAnalyzer.Core.Application;
namespace Gpulse.WCT.DataAnalyzer.Core.Infrastructure.Extensions; namespace Gpulse.WCT.DataAnalyzer.Core.Infrastructure.Extensions;
@@ -58,6 +59,10 @@ public static class ServiceCollectionExtensions
services.AddScoped<CleanService>(); services.AddScoped<CleanService>();
services.AddScoped<AggregationService>(); services.AddScoped<AggregationService>();
// Analysis
services.AddScoped<PlossAnalysisService>();
services.AddScoped<QfodCalibrationService>();
return services; return services;
} }
} }

View File

@@ -0,0 +1,144 @@
using System.CommandLine;
using Gpulse.WCT.DataAnalyzer.Core.Application.Analysis;
namespace Gpulse.WCT.DataAnalyzer.Commands;
/// <summary>
/// 数据分析命令 - 支持 Ploss 余量/delta_p 分析与 Qfod 标定阈值分析
/// </summary>
public class AnalyzeCommand : Command
{
public AnalyzeCommand(
PlossAnalysisService plossAnalysisService,
QfodCalibrationService qfodCalibrationService)
: base("analyze", "Analyze Ploss margin or Qfod calibration threshold from log files")
{
var typeOption = new Option<string>(
"--type",
"Analysis type: ploss or qfod"
)
{
IsRequired = true
};
var fileOption = new Option<string?>(
"--file",
"Ploss: path to a log file to analyze"
);
var pureOption = new Option<string?>(
"--pure",
"Qfod: path to the phone-only charging log file"
);
var foreignOption = new Option<string?>(
"--foreign",
"Qfod: path to the phone + foreign object charging log file"
);
AddOption(typeOption);
AddOption(fileOption);
AddOption(pureOption);
AddOption(foreignOption);
this.SetHandler(async (type, file, pure, foreign) =>
{
type = type.ToLower();
switch (type)
{
case "ploss":
await HandlePlossAsync(plossAnalysisService, file);
break;
case "qfod":
await HandleQfodAsync(qfodCalibrationService, pure, foreign);
break;
default:
Console.WriteLine($"Unknown type: {type}. Use 'ploss' or 'qfod'.");
break;
}
}, typeOption, fileOption, pureOption, foreignOption);
}
private static async Task HandlePlossAsync(PlossAnalysisService service, string? file)
{
if (string.IsNullOrEmpty(file))
{
Console.WriteLine("Please specify --file for Ploss analysis");
return;
}
var report = await service.AnalyzeFileAsync(file);
if (!report.IsSuccess)
{
Console.WriteLine($" [{report.FileName}] FAILED: {report.ErrorMessage}");
return;
}
Console.WriteLine($"\n=== Ploss Analysis: {report.FileName} ===");
Console.WriteLine($"Total: {report.TotalCount} (TwoLine: {report.TwoLineCount}, Legacy: {report.LegacyCount}), FOD: {report.FodCount}, Errors: {report.ErrorCount}");
var normalRatio = Percent(report.NormalAboveCount, report.NormalCount);
var fodRatio = Percent(report.FodAboveCount, report.FodCount);
Console.WriteLine($"Normal margin>2000: {report.NormalAboveCount}/{report.NormalCount} ({normalRatio}%)");
Console.WriteLine($"FOD margin>2000: {report.FodAboveCount}/{report.FodCount} ({fodRatio}%)");
Console.WriteLine($"Identity (delta_p = Tx - Rx - pow_loss - ploss): {report.IdentityHeldCount}/{report.IdentityCheckedCount}");
if (report.FodRows.Count == 0)
{
Console.WriteLine("No FOD rows (Field12 == 1).");
return;
}
Console.WriteLine($"\nFOD rows (delta_p back-solve):");
foreach (var row in report.FodRows)
{
var format = row.IsTwoLine ? "two-line" : "legacy";
Console.WriteLine($" Line {row.LineNumber,-6} [{format}] pow_loss={row.PowLoss?.ToString() ?? "-"} delta_p={row.DeltaP?.ToString() ?? "-"} ploss={row.Ploss} threshold={row.Threshold} margin={row.Margin}");
if (row.IsTwoLine && row.DeltaPPrime.HasValue)
{
Console.WriteLine($" -> delta_p'={row.DeltaPPrime.Value}");
Console.WriteLine($" rewrite: {row.RewrittenHeader}");
}
else
{
Console.WriteLine($" (旧单行格式无 header无法反解 delta_p)");
}
}
}
private static async Task HandleQfodAsync(QfodCalibrationService service, string? pure, string? foreign)
{
if (string.IsNullOrEmpty(pure) || string.IsNullOrEmpty(foreign))
{
Console.WriteLine("Please specify --pure and --foreign for Qfod calibration");
return;
}
var report = await service.CalibrateAsync(pure, foreign);
if (!report.IsSuccess)
{
Console.WriteLine($" FAILED: {report.ErrorMessage}");
return;
}
Console.WriteLine("\n=== Qfod Calibration ===");
PrintQfodFileSummary("Pure phone", report.Pure!);
PrintQfodFileSummary("Foreign object", report.Foreign!);
Console.WriteLine($"\nThreshold = (pureMax + foreignMin) / 2 = ({report.Pure!.Max} + {report.Foreign!.Min}) / 2 = {report.Threshold}");
Console.WriteLine($"Margin: pureMax - threshold = {report.PureMargin} (负为安全), foreignMin - threshold = {report.ForeignMargin} (正为安全)");
}
private static void PrintQfodFileSummary(string label, QfodFileSummary summary)
{
Console.WriteLine($" {label}: {summary.FileName}");
Console.WriteLine($" count={summary.Count} min={summary.Min} max={summary.Max} avg={summary.Average:F1}");
}
private static string Percent(int part, int total)
=> total == 0 ? "0.0" : (part * 100.0 / total).ToString("F1");
}

View File

@@ -7,6 +7,7 @@ using Gpulse.WCT.DataAnalyzer.Commands;
using Gpulse.WCT.DataAnalyzer.Core.Infrastructure.LocalData; using Gpulse.WCT.DataAnalyzer.Core.Infrastructure.LocalData;
using Gpulse.WCT.DataAnalyzer.Core.Infrastructure.Extensions; using Gpulse.WCT.DataAnalyzer.Core.Infrastructure.Extensions;
using Gpulse.WCT.DataAnalyzer.Core.Application; using Gpulse.WCT.DataAnalyzer.Core.Application;
using Gpulse.WCT.DataAnalyzer.Core.Application.Analysis;
namespace Gpulse.WCT.DataAnalyzer; namespace Gpulse.WCT.DataAnalyzer;
@@ -70,6 +71,9 @@ public class Program
rootCommand.AddCommand(new AggregateCommand(provider.GetRequiredService<AggregationService>())); rootCommand.AddCommand(new AggregateCommand(provider.GetRequiredService<AggregationService>()));
rootCommand.AddCommand(new ExportCommand(provider.GetRequiredService<ExportService>())); rootCommand.AddCommand(new ExportCommand(provider.GetRequiredService<ExportService>()));
rootCommand.AddCommand(new CleanCommand(provider.GetRequiredService<CleanService>())); rootCommand.AddCommand(new CleanCommand(provider.GetRequiredService<CleanService>()));
rootCommand.AddCommand(new AnalyzeCommand(
provider.GetRequiredService<PlossAnalysisService>(),
provider.GetRequiredService<QfodCalibrationService>()));
return await rootCommand.InvokeAsync(args); return await rootCommand.InvokeAsync(args);
} }