Muhimman Tsarin Inganci na Kayan Aiki na Koguna na Zangon Yawancin Faduwar Tashin Kirkiro
A nan da shi ne, wanda zaben automation na gudummawa, yawan amfani da load switches a kungiyoyin kungiya ta zama mafi yawa. Amma, abubuwa mai kan mekaniki suna taka waɗannan mutane, tare da muhimmanci ga ƙoƙarin gudummawa da al'amuran ƙungiyar.
Tushen mekanikin ba da kyau shine mafi tsari wajen abubuwan switch. Akwai malamai masu ilimi da suke bincike hanyoyin ƙarin bayyana switchgear, da suke amfani da hukuma kamar yanayin coil current, ƙaramin sinyalin vibration, bayyana travel switch, ƙaramin abubuwan ultrasonic, da kuma ƙaramin temperature da infrared. Yanayin karfi motor current ya yi aiki don circuit breakers da disconnectors, amma ba a yi amfani da ita a matsalolin drive mechanism da take da load switches.
Bayyana a cikin load switches da suke tafiya a fili na musamman ya nuna cewa sinyalin motor energy-storage ya nuna haliyar switch. Matsaloli a cikin mekaniki (kamar jamming spring, rust, gear jamming) suna ɓoye parametos na sinyalin current (amplitude, duration, local peaks). Wannan makaranta ya haɗa da rashin ƙaramin abubuwan fault da take da energy-storage motor rust-jamming, wadanda ana samun a fili na yamma. Ayyuka: 1) Tuntubi ƙaramin sinyalin current, kafa sinyalin da aka kafa zuwa 4 ƙasa, da kuma ƙara each stage. 2) Bincika kafin samun data don sinyalin current daga fagen da suka fi. 3) Bayar da ayyukan algorithm start recording, ƙaramin abubuwan feature, da kuma hukumomin fault. 4) Ƙara da tattaunawa.
1 Ƙaramin Abubuwan Sinyalin Motor Energy-Storage
Load switches suna amfani da DC motors don ƙoƙarin springs don energy storage. A lokacin ƙoƙarin motor, torque da speed da ake fitowa ta ƙarfin stator-circuit current. Tsarin electromagnetic torque da voltage na shunt-excited DC motor suna ƙunshi:
A Equation (1), T na nufin electromagnetic torque; n na nufin rotational speed; Ia na nufin armature current; Ra na nufin resistance na armature circuit, wanda yake da sabon value; Ea na nufin winding induced electromotive force; U na nufin terminal voltage; ΔU na nufin contact voltage drop, wanda yake da sabon value; ϕ na nufin magnetic flux; Ce na nufin electromotive force constant; and CT na nufin torque coefficient. Daga Equation (1), za a iya ƙara:
Daga Equation (2), idan current na load ya ƙece, ya ƙara da demagnetizing effect na armature reaction, kuma ya ƙara da magnetic flux a ƙarfin sabon value, da electromagnetic torque ya ƙara da current na load. Idan current na load ya ƙara, torque ya ƙara, amma speed ya ci gaba. Amma, demagnetizing effect na current na ƙarfin ya ƙara da magnetic flux, wanda ya ƙara da speed. Wanannan effects masu ƙaramin ya ƙara da laifi daɗi a ƙarfin speed na shunt-excited motor. Figure 1 na nuna sinyalin current na DC energy-storage motor a lokacin ƙoƙarin, da aka kafa zuwa 4 ƙasa.Figure 1 shows the typical current waveform of a DC energy - storage motor in operation, divided into 4 stages.
Stage 1 (t0)–(t1): Ƙarfin Motor
A lokacin t0, load switch ya samu signal na closing daga distribution terminal unit, don ƙoƙarin motor control don ƙoƙarin. Sinyalin motor ya ƙara zuwa peak na ƙarfin a (tst), sannan ya ci gaba zuwa stable operation.
Stage 2 (t1)–(t2): Ƙarfin Stable Operation
Motor ya ƙoƙarin gear don idle. A wannan ƙasasshi, motor ya ƙoƙarin stable ta ƙarfin light load, da amplitude na sinyalin motor a (Ia).
Stage 3 (t2)–(t4): Ƙarfin Energy-Storage Spring
Idan compression spring ya ƙoƙarin energy, output torque na motor ya ƙara, ya ƙara zuwa maximum a (t3); a wannan lokaci, sinyalin motor ya ƙara zuwa maximum na ƙasa (Im). Sannan, output torque na motor ya ci gaba ƙara.
Stage 4 (t4)–(t5): Ƙarfin Current Interruption
A (t4), compression spring ya ƙarfin limit switch, tare da power na motor. Sinyalin motor ya ci gaba ƙara har zuwa 0 a (t5), da motor ya ƙoƙarin stop running.
2 Ƙaramin Fault na Jamming na Motor Energy-Storage
2.1 Simulation Fault & Samun Data
An yi test na jamming fault a load switch daga factory na electrical equipment (scenario a Fig. 2(a)). Ba a ƙarfe switch, a lokacin ƙarfin stable da ƙarfin energy-storage, an yi reverse locked-rotor forces don simulate gear/spring jamming. Kafin samun current (Fig. 2(b)) an yi amfani da ARM STM32F103 chip don samun signals daga HSTS016L Hall current transformer (DC input: 0–30A). Saboda signal na opening ba da target current waveform, wannan bincike ta haɗa da closing current signal.
2.2 Algorithm Start Waveform Recording
Daga Figure 1, sinyalin effective signal ya ƙarfin time window t0 zuwa t5， da ke ƙasa 4 ƙasa da ƙaramin current changes. Da ƙarin, akwai ƙaramin sabon abubuwan signal amplitudes daga fagen da suke fi. Saboda haka, amfani da simple current amplitude threshold as the start criterion for signal waveform recording is clearly inappropriate. Therefore, this study adopts the current change rate Kt within a unit time window and the mean value Imean as the start criteria to achieve effective waveform recording.Current change rate of the unit time window:
Mean current of each time window:
A Equations (3) and (4), Ii na nufin current signal; M na nufin number of sampling points in the unit time window; Δ t na nufin time length of the unit time window, and Δ t = 0.02s in this paper; I(1) na nufin first sampling point in the unit time window.
2.3 Ƙaramin Abubuwan Time-Domain
Don ƙara fault na jamming na motor energy-storage, expressive information na curve ya ƙara don ƙaramin abubuwan time-domain indicators. Kurtosis K can characterize the smoothness of the current signal; root mean square Irms can characterize the average energy of the current signal; skewness sk is a measure of the direction and degree of skewness of the statistical data distribution; form factor sh and the peak factor C are used to characterize the extreme degree of the current peak in the waveform.
The Random Forest (RF)classification algorithm integrates multiple decision trees. Its output category is determined by the mode of individual decision-tree categories, featuring high accuracy, good tolerance for abnormal data, and low overfitting risk.
2.4 Random Forest Algorithm
RF relies on Bootstrap sampling (with-replacement sampling to form n sample sets from the original dataset) and Bagging voting. Bagging generates n training sets via Bootstrap, each training an independent weak classifier. Final decisions come from voting on weak-classifier outputs, with the majority vote as the result.
RF uses CART decision trees (binary trees splitting top-down from the root, minimizing the Gini index for splits, formula (5)). Per Liu Min et al. 100 decision trees optimize classification performance. Thus, this study uses 100 CART trees for the random forest.
3 Case Analysis
3.1 Feature Selection
Gini index in the random forest is used to evaluate the importance of each feature. The results are shown in Figure 3, where the ordinate represents the proportional coefficient. It can be seen that four feature quantities, namely the peak factor C, skewness sk, root mean square Irms, and kurtosis K, are highly important and can effectively characterize the differences in different states of the load switch. The four feature quantities, including the form factor sh, maximum starting current Ist, motor operating time t, and Tm, are of low importance. Therefore, this study selects C, sk, Irms, and K as the feature vectors.
3.2 Random Forest Diagnosis Results
The RF algorithm classifies two load-switch states (normal/jammed) using 300 samples per state for training (total 600) and 30 samples for testing. The confusion matrix (Figure 4) shows perfect normal-state identification, 97% accuracy for jamming, and 98.33% average classification accuracy.
3.3 Comparison of Different Classification Algorithms
To test the performance of the random forest classifier, a Support Vector Machine (SVM) and an Extreme Learning Machine (ELM) are trained simultaneously for comparison. The test results are shown in Table 1.
From Table 1, among the three classifiers, the Random Forest (RF) algorithm takes a relatively long diagnosis time of 6.9 ms for test set samples. In terms of accuracy, the Support Vector Machine (SVM) achieves 95% for two operating states, lower than RF. Due to random hidden-layer weights, the Extreme Learning Machine (ELM) has accuracy fluctuating between 85% - 96.67% and poorer robustness than RF. Thus, the RF algorithm used has high accuracy and good robustness.
4 Conclusion
This paper proposes a load-switch mechanical fault detection method using energy-storage motor current time-domain features and the Random Forest (RF) algorithm. It extracts representative time-domain features from motor current waveforms and uses an RF classifier for state identification. The proposed recording-wave start criterion effectively acquires motor current signals. Leveraging the Gini index in RF, it evaluates feature importance and selects four key features (peak factor, skewness, root mean square, kurtosis) to characterize load-switch states. Experiments show the method effectively identifies motor jamming states with 98.33% accuracy.
