Direct answer

To discuss a "before-and-after comparison" in harmonic treatment, the first step is not choosing a filter but measuring the pre-treatment harmonic state accurately and locating the pollution source accurately. The product knowledge base provides capability in three parts: on the acquisition side, the power-quality monitor provides phase and harmonic monitoring, covering harmonics of order 2 to 31 at an accuracy of ±1%; on the analysis side, the power-quality check-up board of the Qianzhi engine provides harmonic sub-models covering order 2 to 50 and total harmonic distortion, together with a harmonic fingerprint library for matching pollution sources; and on the responsibility side, the harmonic responsibility division capability of the Tianyan engine quantifies the harmonic contribution of the user side and the grid side. One boundary must be written first: the product knowledge base does not specify the measurement period, the target order acceptance threshold or the pass criterion for verifying the effect of harmonic treatment, so this article describes only the available monitoring, matching and attribution capabilities, and infers no specific acceptance-standard value.

1. Why "before-and-after" must first solve comparability

The effect of harmonic treatment is usually compared through the harmonic orders and total harmonic distortion before and after treatment. For the comparison to hold, however, the two measurements must share the same definition: same acquisition location, same harmonic order range, and comparable load conditions. If the comparison rests only on complaints before and subjective impressions after, it lacks a common baseline.

The support the product knowledge base gives here is layered capability: the power-quality monitor for acquisition, the Qianzhi engine for analysis and fingerprint matching, and the Tianyan engine for responsibility quantification. Only by linking these three parts is the before-and-after comparison given a traceable data basis. This article sets out the four steps of acquisition, analysis, tracing and attribution in turn, while respecting the acceptance boundary the knowledge base does not give.

2. Acquisition: what the power-quality monitor measures

The acquisition role is taken by the power-quality monitor. The product knowledge base records that the power-quality monitor (ESE-22111-R) adds harmonic monitoring on top of phase monitoring, with a harmonic coverage of order 2 to 31 and an accuracy of ±1%; it provides RS485 (Modbus) communication, two switching-value inputs and one relay output.

These are the parameters given verbatim by the product knowledge base. What can be confirmed is that the product provides both phase and harmonic measurement and has basic switching-value input and relay output interfaces. The product knowledge base gives no sampling rate, voltage or current range, limit alarm threshold or synchronisation method, so this article adds none of these unlisted figures. If this product is used for before-and-after acquisition, the two measurements must use the same model and the same wiring location, otherwise the definitions are not comparable.

3. Analysis: the harmonic sub-models of the Qianzhi engine

The analysis role falls on the power-quality check-up board of the Qianzhi engine. The product knowledge base records that M06 to M12 of this board include harmonic sub-models covering order 2 to 50 harmonics and total harmonic distortion (THD). This range is wider than the order 2 to 31 of a single acquisition monitor, showing the analysis layer can take harmonic data from different acquisition products.

The value of the harmonic sub-model is to organise the raw waveform into comparable indicators: which orders stand out, how much each contains, and what level the total harmonic distortion is at. Record one set before treatment and one set after, and the two order spectra and total harmonic distortions can be compared. Note that the product knowledge base gives only the order range and indicators the sub-model covers, not its algorithm implementation or decision thresholds, and this article does not develop the internal logic.

4. Tracing: how the fingerprint library locates the pollution source

The most time-consuming step in harmonic treatment is often not acquisition but deciding where the pollution comes from. The product knowledge base records that the harmonic fingerprint library contains 14 classes of equipment fingerprints, such as FP-01 three-phase rectifier, FP-03 six-pulse frequency converter, FP-05 uninterruptible power supply, FP-06 charging pile and FP-12 photovoltaic inverter. Fingerprint matching uses a condition of cosine similarity greater than 0.85 and, according to the record, can lock the pollution source within 2 hours, whereas the traditional method takes weeks.

The meaning of this record is to move treatment forward from "adding equipment afterwards" to "locating at the source": first confirm which class of device produces the characteristic harmonics, then decide the priority of the treatment target. Note that the product knowledge base gives the number of fingerprint classes, the matching threshold and the lock-time definition, not the complete list, the similarity calculation method or the misjudgement rate, and this article adds none of these. If the same pollution source can be locked before and after treatment, the attribution of responsibility and effect has a basis.

5. Attribution: how the user side and the grid side are divided

Who bears the investment in treatment depends on whether the harmonic contribution comes from the user side or the grid side. The product knowledge base records that Q-01 harmonic responsibility division of the Tianyan engine is based on the IEEE 1459 standard and is used to quantify the harmonic contribution of the user side and the grid side; it also records a field case in which the rectification cost fell from 80 ten-thousand yuan to 28 ten-thousand yuan.

This shows that responsibility division can provide a quantitative basis for attributing treatment investment. This article cites only the result definition of that case and does not add the project name, time, measurement layout or cost-sharing rule. Note that the product knowledge base gives no complete process or decision threshold for responsibility determination, so attribution should be treated as an evidence-gathering process, not a one-click conclusion.

6. Selection combination: how monitoring, fingerprint and attribution fit together

In the selection comparison of the product knowledge base, the combination for harmonic tracing and responsibility division corresponds to Qianzhi M06, the fingerprint library and Tianyan Q-01 (IEEE 1459); for power-quality and harmonic special-treatment scenarios it recommends the power-quality monitor (ESE) or the multi-parameter electrical intelligent controller (power quality type, corresponding to a model such as SFE-11111-R, with Tianyan engine harmonic analysis).

These two combinations show that harmonic work cannot be completed by a single product: acquisition needs power-quality monitoring capability, analysis needs harmonic sub-models, location needs the fingerprint library, and attribution needs responsibility division. The product knowledge base gives only the combination relationships, not the wiring method, data flow or responsibility-sharing ratio, and this article does not develop them into a specific scheme.

7. Common misreadings and reading order

The first misreading is to measure only total harmonic distortion without the order spectrum, so that before and after cannot correspond. The second is to mix the order 2 to 31 of the acquisition side with the order 2 to 50 of the analysis side as one definition, ignoring that they come from different links. The third is to skip fingerprint matching and add filter equipment directly, leaving the treatment target unclear. The fourth is to treat responsibility division as automatic determination and ignore that the knowledge base gives no decision threshold. The fifth is to treat one case's cost change as a general effect promise.

The corresponding reading order is: first confirm the acquisition product and harmonic order range, then the coverage of the analysis sub-model, then use the fingerprint library to locate the pollution source, and finally use the responsibility-division capability to define the user-side and grid-side contributions.

Scope and limitations

First, the factual basis of this article is the product knowledge base, and all product parameters are limited to what it lists. Second, the product knowledge base does not specify the measurement period, target order acceptance threshold or pass criterion for verifying the effect of harmonic treatment, and this article infers no specific acceptance-standard value. Third, the order 2 to 31 harmonic monitoring and ±1% accuracy, RS485 (Modbus), two switching-value inputs and one relay output of the power-quality monitor (ESE-22111-R) are cited as listed, and this article adds no sampling rate or range. Fourth, the order 2 to 50 and total harmonic distortion of the Qianzhi engine harmonic sub-model, and the 14 equipment fingerprint classes, cosine similarity greater than 0.85 and 2-hour lock of the harmonic fingerprint library, are cited as listed, and this article adds no complete list or misjudgement rate. Fifth, the Q-01 responsibility division and IEEE 1459 basis of the Tianyan engine, and the case result of a fall from 80 ten-thousand yuan to 28 ten-thousand yuan, are cited as listed, and this article adds no cost-sharing rule. Sixth, this article promises no treatment effect or acceptance conclusion; the actual situation is subject to the latest product material and formal documents.