Digital sensor system

Abstract

The invention discloses a digital sensor system, which relates to the technical field of power system monitoring and comprises an acquisition unit, a collection unit and a master station. By adopting a time-frequency domain joint feature analysis algorithm, the system can accurately collect line current and electric field signals and perform short-time Fourier transform and feature extraction on the line current and electric field signals, so that a fault feature matrix is constructed, and the algorithm can accurately judge interphase short circuit and single-phase grounding faults and can also accurately judge the fault characteristics of the line. According to the method, fault types can be distinguished according to energy distribution characteristics in a characteristic matrix, the accuracy of fault detection is greatly improved, meanwhile, the system has the characteristic of low power consumption, long-service-life operation can be guaranteed under the condition that no external power supply exists, and the real-time performance and continuity of fault detection are ensured.

Description

Technical Field

[0001] The invention relates to the technical field of power system monitoring, in particular to a digital sensor system. Background Art

[0002] With the continuous development of power systems and the increasing demand for intelligence, real-time monitoring of power grid status and rapid fault location have become important links in power system operation and maintenance. As an advanced technical means, digital sensor systems are gradually being applied to the monitoring field of power systems to improve system reliability and safety.

[0003] Traditional power system monitoring technology has many shortcomings. Traditional sensors have limitations in measurement accuracy and stability, and it is difficult to meet the high-precision monitoring needs in complex and changeable line operation environments. The lack of comprehensive line operation data support leads to inaccurate fault location and cause analysis.

[0004] In summary, traditional power system monitoring technology has obvious deficiencies in measurement accuracy, data acquisition and processing, communication methods, equipment size and power consumption, and it is difficult to meet the high requirements of modern power systems for real-time monitoring and rapid fault location. Therefore, it is particularly important to develop a digital sensor system. Summary of the invention

[0005] The purpose of the present invention is to make up for the deficiencies of the prior art and to provide a digital sensor system. By adopting a time-frequency domain joint feature analysis algorithm, the system can accurately collect line current and electric field signals and perform short-time Fourier transform and feature extraction on them, thereby constructing a fault feature matrix. The system can also distinguish fault types according to the energy distribution characteristics in the feature matrix, thereby greatly improving the accuracy of fault detection.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a digital sensor system, the system includes a collection unit, a collection unit and a main station;

[0007] The acquisition unit collects fault indication information and line operation information, and its measurement accuracy satisfies that when the line current is in the range of OA-100A, the measurement error is ±1A, and when the line current is in the range of 100A-630A, the measurement accuracy is ±1. The measurement accuracy is achieved based on an adaptive weighted fusion measurement algorithm, and the algorithm formula is: Among them, I 精准测量 To accurately measure the current value, I i is the current value collected by the i-th sensor in the acquisition unit, a i and b i is the calibration coefficient of the i-th sensor, n is the number of sensors involved in current measurement in the acquisition unit, and the calibration coefficient ai and b i In a laboratory environment, a current signal with a known precise current value is input to each sensor, the sensor output value is measured, and then the least squares method is used to fit the value, so that the error between the sensor output value and the precise current value is minimized. The number of sensors n in the acquisition unit is determined according to the complexity of the line. For complex lines, n is 5-8, and for relatively simple lines, n is 3-5. At the same time, the acquisition unit has low power consumption characteristics, with standby power consumption less than 40uA and average operating power consumption less than 90uA. It has a built-in 3.6V, 7.2Ah capacity lithium battery to ensure long-life operation without external power supply. Its low power design adopts a dynamic power management algorithm to allocate power according to the working status and task priority of the acquisition unit. The algorithm formula is:

[0008] Where P 分配 is the power allocated to the current task, P 总 is the total power supply, T j is the estimated execution time of the current task, T 总 Total task cycle time, W j is the priority weight of the current task, m is the number of tasks that exist at the same time, and the priority weight W j The acquisition unit is designed to be lightweight and reliable according to the impact of the task on the real-time performance and accuracy of the system. It has been verified through multiple rounds of high and low temperature experiments. Its fault detection is based on an innovative joint feature analysis algorithm in the time and frequency domains. First, the collected current and electric field signals are subjected to short-time Fourier transform. The formula is: Among them, x(τ) is the input signal, w(τ-t) is the time window function, t is the time variable, and f is the frequency variable. Then, the STFT results are extracted to construct the fault feature matrix. For short-circuit faults, the fault is judged based on the high-frequency time-frequency domain energy distribution characteristics in the feature matrix. For grounding faults, the judgment is made based on the correlation between the low-frequency time-frequency domain energy change and the high-frequency time-frequency domain energy change. The acquisition unit can provide line current, ground electric field, whether the fault state is energized, and auxiliary information such as ambient temperature and battery voltage. An ultra-bright LED is used for 360° omnidirectional visual status indication. After the line fault is eliminated and power supply is restored, it can be automatically reset, or it can be automatically reset at a scheduled time or remotely set and reset by wireless. The scheduled automatic reset time can be set from 0 to 48 hours, and the default is 24 hours. The reset time accuracy is ±1%;

[0009] The collection unit integrates data collection, data processing and data transmission functions, collects information from AC two-phase collection units through 433Mhz radio frequency, and sends the data to the main station through GPRS. It has a built-in large-capacity, long-life battery, and can continuously draw power within the line current range of 0-650A. When the line has power, the supercapacitor is charged to meet the operating energy consumption. The power control is based on an innovative intelligent current tracking algorithm, and the algorithm formula is: Among them I 取电 is the current drawn, I 线 is the line current, ΔI1 is the current increment dynamically adjusted according to the line current fluctuation, T c is the power-taking control cycle, T s is the system sampling period. The current increment ΔI1 is predicted based on the historical fluctuation data of the line current and the real-time monitoring data. The adaptive filtering algorithm is used for calculation. The size of the collection unit is 120mmx126mmx178mm, the weight is <600g, the protection level is IP68, the clamping force is 5 times its own weight in the vertical direction and 50N in the line direction without displacement, the loading and unloading life is >50 times without damage, the mechanical strength reaches vibration level 1, it can be tilted and dropped 1 meter, the short-circuit current impact tolerance is 40kA / 4S, the fire hazard level is level 5, the radio frequency electromagnetic field radiation immunity is level 3, the adjacent interference test is 300mm, the operating life is >5 years, and the electrical life is >2000 times;

[0010] The master station is used to receive the fault information sent by the collection unit, and in combination with the GIS system, displays the accurate fault point and the current change when the fault occurs in real time, and gives an audible and visual alarm, and notifies the relevant staff by SMS.

[0011] Furthermore, the arrangement of sensors in the acquisition unit is optimized based on the three-dimensional distribution model of the line electric field and magnetic field. The three-dimensional distribution model of the electric field and magnetic field around the line is established by the finite element analysis method. The optimal position of the sensor is determined according to the gradient change of the electric field strength and the magnetic field strength in the model. In the area where the electric field strength changes dramatically, sensors sensitive to the electric field are preferentially arranged, and in the area where the magnetic field strength changes dramatically, sensors sensitive to the magnetic field are preferentially arranged, so as to improve the measurement accuracy and the sensitivity of fault detection.

[0012] Furthermore, the lithium battery management system of the acquisition unit adopts a remaining power prediction algorithm, which combines the internal resistance change, temperature change and charge and discharge history records of the battery to predict the remaining power. The formula is: Where Q 剩余 To predict the remaining power, Q 初始 is the initial battery charge, I 安际 is the actual current, R t is the current battery internal resistance, R0 is the initial battery internal resistance, T tis the current battery temperature, T0 is the initial battery temperature, is the change in charge during the i-th charge and discharge process, p is the number of charge and discharge times, the battery internal resistance is calculated using Ohm's law by real-time measuring the voltage across the battery and the current flowing through the battery, and the battery temperature is measured by a built-in temperature sensor.

[0013] Furthermore, the electromagnetic compatibility protection of the acquisition unit adopts an adaptive shielding optimization algorithm, which adjusts the parameters of the shielding layer in real time according to the frequency and intensity of the external electromagnetic interference. s and equivalent capacitance C s The adjustment formula is: Where L0 and C0 are the initial equivalent inductance and capacitance of the shielding layer, f 干扰 is the external interference frequency, f0 is the system operating frequency, I 干扰 is the external interference intensity, I0 is the normal working current of the system, and the shielding layer parameters are adjusted to achieve the best shielding effect under different interference environments.

[0014] Furthermore, in the fault detection algorithm of the acquisition unit, the time window function of the short-time Fourier transform is adaptively selected according to the time-varying characteristics of the signal. For the fast-changing fault transient signal, a shorter time window is selected to improve the time resolution. For the relatively slowly changing steady-state signal, a longer time window is selected to improve the frequency resolution. The time window length T w The adjustment formula is: Where T min and T max is the minimum and maximum time window length, Δf is the signal frequency change rate, f max is the highest frequency of the signal.

[0015] Furthermore, the lithium battery charging management of the collection unit adopts a multi-stage dynamic charging strategy. In the initial stage of charging, a constant current charging method is adopted, and the charging current is determined according to the initial battery power and temperature. The formula is: Among them I max is the maximum allowable charging current, Q is the full charge of the battery. When the battery charge reaches a certain threshold (such as 50%), it switches to constant voltage charging mode. The charging voltage is dynamically adjusted according to the battery characteristics and temperature. The adjustment formula is: Where V 标准 is the standard charging voltage, V max is the maximum allowable charging voltage, T max This is the maximum temperature allowed for the battery.

[0016] Furthermore, the wireless communication module of the aggregation unit adopts an adaptive coding and modulation algorithm based on signal quality assessment. This algorithm selects the coding and modulation mode according to the signal-to-noise ratio (SNR), bit error rate (BER), and channel fading of the received signal. The basis for the selection of the coding and modulation mode is as follows: when SNR≥SNR1 and BER≤BER1, a high-order modulation and low-redundancy coding mode are selected to improve the communication rate; when SNR2≤SNR<SNR1 or BER1<BER≤BER2, a moderate modulation and coding mode are selected to balance the communication rate and reliability; when SNR<SNR2 or BER>BER2, a low-order modulation and high-redundancy coding mode are selected to ensure communication reliability, where SNR1, SNR2, BER1, and BER2 are thresholds preset according to system requirements.

[0017] Furthermore, the wire clamping structure of the aggregation unit adopts an intelligent adaptive clamping mechanism. This mechanism uses an in-built pressure sensor to monitor the clamping force in real time. The adjustment of the clamping force is based on the material properties, wire diameter, and temperature of the wire. The clamping force F 抱紧 is calculated by the formula: where F0 is the initial clamping force, D is the current wire diameter, D0 is the standard wire diameter, E t is the elastic modulus of the current wire material, E0 is the elastic modulus of the standard wire material, k is the temperature coefficient, T is the current wire temperature, and T0 is the standard temperature. By adjusting the clamping force in real time, it is ensured that it can be stably and reliably installed on the wire under different working conditions, while avoiding damage to the wire.

[0018] Furthermore, the fault location algorithm of the master station adopts an intelligent location method based on machine learning. A neural network model is trained using historical fault data and real-time monitoring data. The inputs of the model include line current, voltage, power factor, and fault indicator status information, and the output is the predicted value of the fault location. The structure and parameters of the neural network model are optimized by the backpropagation algorithm, and the connection weights between neurons are adjusted according to the prediction error. During the training process, a cross-validation method is adopted to improve the generalization ability of the model, ensuring that the fault location can be accurately predicted under different line conditions.

[0019] Furthermore, the wireless communication between the acquisition unit and the aggregation unit adopts a dynamic key negotiation encryption protocol. Before each communication, both parties generate a temporary session key according to the pre-shared key seed and random numbers. The key generation formula is: K 会话 =Hash(K 种子 ||R3||R40, where Hash is a hash function, K 种子 is the shared key seed, and R3 and R4 are random numbers generated by both parties. By dynamically generating the session key, the security of communication is improved, preventing information leakage caused by the key being cracked.

[0020] Compared with the prior art, a digital sensor system has the following beneficial effects:

[0021] 1. By adopting a time-frequency domain joint feature analysis algorithm, the present invention enables the system to accurately collect line current and electric field signals, and perform short-time Fourier transform and feature extraction on them, thereby constructing a fault feature matrix. This algorithm can not only accurately determine phase-to-phase short circuit and single-phase grounding faults, but also distinguish the fault types according to the energy distribution characteristics in the feature matrix, greatly improving the accuracy of fault detection. At the same time, the system has low power consumption characteristics and can ensure long-life operation without external power supply, ensuring the real-time and continuity of fault detection.

[0022] 2. The present invention can adaptively adjust the number of sensors according to the complexity of the line through the acquisition unit, and improve the measurement accuracy through the adaptive weighted fusion measurement algorithm. The collection unit integrates data acquisition, fault data processing and data transmission functions, realizes continuous power supply through the intelligent current tracking algorithm, and optimizes the performance of the wireless communication module through the adaptive coding modulation algorithm based on signal quality evaluation. The main station uses the intelligent positioning method of machine learning and combines the GIS system to display the accurate fault point and the current change when the fault occurs in real time, providing strong support for fault handling and power supply restoration. The whole system realizes intelligent data acquisition, processing and transmission, and improves the reliability and safety of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 A digital sensor system operation flow chart. DETAILED DESCRIPTION

[0025] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] Embodiment 1

[0027] This embodiment describes that in rural areas, the power grid is widely distributed and the lines are complex. It is difficult for traditional monitoring methods to locate fault points in a timely and accurate manner, which brings difficulties to fault handling. In addition, rural areas often lack efficient communication infrastructure, making the transmission and processing of fault information even more difficult.

[0028] Collecting units are deployed on the key lines and fault-prone areas of the rural power grid. The installation locations of the collecting units should take into account topographical and climatic factors to ensure that the sensors can work stably for a long time. The collecting units adopt a time-frequency domain joint feature analysis algorithm, which can accurately detect phase-to-phase short circuits and single-phase grounding faults and construct a fault feature matrix. Where x(τ0 is the input signal, w(τ - t) is the time window function, t is the time variable, f is the frequency variable, to improve the accuracy of fault detection. The collecting units also have the ability of real-time data preprocessing, which can preliminarily screen and filter invalid data to reduce the processing burden of the master station.

[0029] The aggregation unit communicates wirelessly with the collecting units via 433Mhz radio frequency, overcoming the challenges of complex terrain and difficult signal transmission in rural areas. Considering the communication coverage problem in rural areas, relay nodes can be added or a wireless Mesh network structure can be adopted to improve the communication coverage and reliability. The aggregation unit adopts an adaptive coding and modulation algorithm based on signal quality assessment. This algorithm selects the coding and modulation method according to the signal-to-noise ratio SNR, bit error rate BER and channel fading of the received signal. The selection basis of the coding and modulation method is as follows: when SNR≥SNR1 and BER≤BER1, a high-order modulation and low-redundancy coding method is selected to improve the communication rate; when SNR2≤SNR<SNR1 or BER1<BER≤BER2, a moderate modulation and coding method is selected to balance the communication rate and reliability; when SNR<SNR2 or BER>BER2, a low-order modulation and high-redundancy coding method is selected to ensure communication reliability. Among them, SNR1, SNR2, BER1 and BER2 are thresholds preset according to system requirements. The coding and modulation method is selected according to the signal-to-noise ratio and bit error rate of the received signal to ensure the reliability and stability of data transmission. In the case of poor signal quality, a low-rate and high-reliability coding method is adopted; in the case of good signal quality, a high-rate and high-efficiency coding method is adopted.

[0030] The master station receives the fault information sent by the collection unit, combines with the GIS system to display the location of the fault point in real time, and gives an audible and visual alarm. The GIS system should support map zooming, panning and rotation operations to facilitate staff to view the fault point and its surrounding environment. The master station notifies relevant staff through text messages, including nearby maintenance personnel and power grid dispatchers, to ensure that the fault can be handled in time. The master station should also have a dispatching function to automatically assign maintenance tasks according to the type and severity of the fault, and track the progress and results of the maintenance. The master station should save all fault information and maintenance records for subsequent analysis and improvement. The master station can also use big data and artificial intelligence technology to mine and analyze fault data, discover potential fault patterns and trends, and provide a scientific basis for preventive maintenance.

[0031] Embodiment 2

[0032] This embodiment describes a busy city center area where the power grid system is complex and critical. Traditional monitoring methods have obvious deficiencies in rapid fault location and high-precision measurement, and are difficult to meet the high requirements of modern power systems for real-time monitoring and rapid fault location.

[0033] Collection units are deployed at key line nodes in the urban power grid, including but not limited to lines at the intersection of main roads, power supply lines in important commercial areas, and main power supply lines in residential areas. Each collection unit adaptively adjusts the number of sensors according to the complexity of the line. For complex lines, such as lines at the intersection of main roads, the number of sensors is set to 5-8 to ensure high-precision measurement. For relatively simple lines, such as the main power supply lines in residential areas, the number of sensors is set to 3-5. The collection unit adopts a lightweight and reliable design, with a high level of electromagnetic compatibility protection capability, and can adapt to the complex electromagnetic environment of the city center. At the same time, the collection unit has low power consumption characteristics, with standby power consumption less than 40uA and average operating power consumption less than 90uA. It has a built-in 3.6V, 7.2Ah capacity lithium battery to ensure long-life operation without external power supply.

[0034] The collection unit is integrated near the substation of the urban power grid to facilitate the reception and processing of data from the collection unit. It collects information from surrounding collection units through 433Mhz radio frequency and sends the data to the main station through GPRS. This method ensures the real-time and reliability of the data. The collection unit has a built-in large-capacity, long-life lithium battery and uses an intelligent current tracking algorithm to continuously draw power within the line current range of 0-650A. Among them I 取电 is the current drawn, I 线 is the line current, ΔI1 is the current increment dynamically adjusted according to the line current fluctuation, T c is the power-taking control cycle, T sis the system sampling period, and the current increment ΔI1 is predicted based on the historical fluctuation data of the line current and the real-time monitoring data. When the line is energized, the supercapacitor is charged to meet the operating energy consumption to ensure continuous operation when the grid is powered. The collection unit has a size of 120mmx126mmx178mm, weighs less than 600g, has a protection level of IP68, and has high mechanical strength and anti-interference capabilities.

[0035] The master station receives fault information and line operation data sent by the aggregation unit, including line current, ground electric field, fault status, whether the working condition is energized, ambient temperature, and battery voltage auxiliary information. Combined with the GIS system, it displays the accurate fault point and current changes when the fault occurs in real time, which helps the staff to quickly locate the fault location and take corresponding measures to repair it. The master station adopts an intelligent positioning method based on machine learning, and uses historical fault data and real-time monitoring data to train the neural network model. The input of the model includes line current, voltage, power factor, and fault indicator status information, and the output is the predicted value of the fault location. This method improves the accuracy and efficiency of fault location. When a fault is detected, the master station will give an audible and visual alarm and notify the relevant staff by text message, which helps the staff to respond and handle the fault in a timely manner.

[0036] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A digital sensor system, characterized in that: The system includes a collection unit, an aggregation unit and a master station: The acquisition unit collects fault indication information and line operation information, and its measurement accuracy satisfies that when the line current is in the range of OA-100A, the measurement error is ±1A, and when the line current is in the range of 100A-630A, the measurement accuracy is ±1. The measurement accuracy is achieved based on an adaptive weighted fusion measurement algorithm, and the algorithm formula is: Among them, I 精准测量 To accurately measure the current value, I i is the current value collected by the i-th sensor in the acquisition unit, a i and b i is the calibration coefficient of the i-th sensor, n is the number of sensors involved in current measurement in the acquisition unit, and the calibration coefficient a i and b i In a laboratory environment, a current signal with a known precise current value is input to each sensor, the sensor output value is measured, and then the least squares method is used to fit the value, so that the error between the sensor output value and the precise current value is minimized. The number of sensors n in the acquisition unit is determined according to the complexity of the line. For complex lines, n is 5-8, and for relatively simple lines, n is 3-5. At the same time, the acquisition unit has low power consumption characteristics, with standby power consumption less than 40uA and average operating power consumption less than 90uA. It has a built-in 3.6V, 7.2Ah capacity lithium battery to ensure long-life operation without external power supply. Its low power design adopts a dynamic power management algorithm to allocate power according to the working status and task priority of the acquisition unit. The algorithm formula is: Where P 分配 is the power allocated to the current task, P 总 is the total power supply, T j is the estimated execution time of the current task, T 总 Total task cycle time, W j is the priority weight of the current task, m is the number of tasks that exist at the same time, and the priority weight W j The acquisition unit is designed to be lightweight and reliable according to the impact of the task on the real-time performance and accuracy of the system. It has been verified through multiple rounds of high and low temperature experiments. Its fault detection is based on an innovative joint feature analysis algorithm in the time and frequency domains. First, the collected current and electric field signals are subjected to short-time Fourier transform. The formula is: Among them, x(τ0 is the input signal, w(τ-t0 is the time window function, t is the time variable, and f is the frequency variable. Then, the STFT results are feature extracted to construct the fault feature matrix. For short-circuit faults, the fault is judged based on the high-frequency time-frequency domain energy distribution characteristics in the feature matrix. For grounding faults, the judgment is made based on the correlation between the low-frequency time-frequency domain energy change and the high-frequency time-frequency domain energy change. The acquisition unit can provide line current, ground electric field, whether the fault state is energized, and auxiliary information such as ambient temperature and battery voltage. An ultra-bright LED is used for 360° omnidirectional visual status indication. After the line fault is eliminated and the power supply is restored, it can be automatically reset, or it can be automatically reset at a scheduled time or remotely set and reset by wireless. The scheduled automatic reset time can be set from 0 to 48 hours, the default is 24 hours, and the reset time accuracy is ±1%; The collection unit integrates data collection, data processing and data transmission functions, collects AC two-phase collection unit information through 433Mhz radio frequency, and sends the data to the main station through GPRS. It has a built-in large-capacity, long-life battery, and can continuously draw power within the line current range of 0-650A. When the line has power, the supercapacitor is charged to meet the energy consumption of the operation. The power control is based on the intelligent current tracking algorithm, and the algorithm formula is: Among them I 取电 is the current drawn, I 线 is the line current, ΔI1 is the current increment dynamically adjusted according to the line current fluctuation, T c is the power-taking control cycle, T s is the system sampling period. The current increment ΔI1 is predicted based on the historical fluctuation data of the line current and the real-time monitoring data. The adaptive filtering algorithm is used for calculation. The size of the collection unit is 120mmx126mmx178mm, the weight is <600g, the protection level is IP68, the clamping force is 5 times its own weight in the vertical direction and 50N in the line direction without displacement, the loading and unloading life is >50 times without damage, the mechanical strength reaches vibration level 1, it can be tilted and dropped 1 meter, the short-circuit current impact tolerance is 40kA / 4S, the fire hazard level is level 5, the radio frequency electromagnetic field radiation immunity is level 3, the adjacent interference test is 300mm, the operating life is >5 years, and the electrical life is >2000 times; The master station: is used to receive the fault information sent by the aggregation unit, combine with the GIS system to display the accurate fault point and the current change situation when the fault occurs in real time, and at the same time give an audible and visual alarm and send a text message to notify the relevant staff.

2. A digital sensor system according to claim 1, characterized in that: The arrangement of the sensors in the collection unit is optimized based on the three-dimensional distribution model of the line electric field and magnetic field. The three-dimensional distribution model of the electric field and magnetic field around the line is established by the finite element analysis method. The optimal position of the sensors is determined according to the gradient changes of the electric field strength and magnetic field strength in the model. In the area where the electric field strength changes violently, the sensors sensitive to the electric field are preferentially arranged. In the area where the magnetic field strength changes violently, the sensors sensitive to the magnetic field are preferentially arranged.

3. A digital sensor system according to claim 1, characterized in that: The lithium battery management system of the acquisition unit adopts a remaining power prediction algorithm, which combines the internal resistance change, temperature change and charge and discharge history of the battery to predict the remaining power. The formula is: Where Q 剩余 To predict the remaining power, Q 初始 is the initial battery charge, I 安际 is the actual current, R t is the current battery internal resistance, R0 is the initial battery internal resistance, T t is the current battery temperature, T0 is the initial battery temperature, is the change in charge during the i-th charge and discharge process, p is the number of charge and discharge times, the battery internal resistance is calculated using Ohm's law by real-time measuring the voltage across the battery and the current flowing through the battery, and the battery temperature is measured by a built-in temperature sensor.

4. A digital sensor system according to claim 1, characterized in that: The electromagnetic compatibility protection of the acquisition unit adopts an adaptive shielding optimization algorithm, which adjusts the parameters of the shielding layer in real time according to the frequency and intensity of the external electromagnetic interference. s And the equivalent capacitance C s The adjustment formula is: Where L0 and C0 are the initial equivalent inductance and capacitance of the shielding layer, f 干扰 is the external interference frequency, f0 is the system operating frequency, I 干扰 is the external interference intensity, I0 is the normal working current of the system, and the shielding layer parameters are adjusted to achieve the best shielding effect under different interference environments.

5. A digital sensor system according to claim 1, characterized in that: In the fault detection algorithm of the acquisition unit, the time window function of the short-time Fourier transform is adaptively selected according to the time-varying characteristics of the signal. For a fast-changing fault transient signal, a shorter time window is selected, and for a relatively slow-changing steady-state signal, a longer time window is selected. The time window length T w The adjustment formula is: Where T min and T max is the minimum and maximum time window length, Δf is the signal frequency change rate, f max is the highest frequency of the signal.

6. A digital sensor system according to claim 1, characterized in that: The lithium battery charging management of the collection unit adopts a multi-stage dynamic charging strategy. In the initial stage of charging, a constant current charging method is adopted. The charging current is determined according to the initial battery power and temperature. The formula is: Among them I max is the maximum allowable charging current, Q is the full charge of the battery. When the battery charge reaches a certain threshold (such as 50%), it switches to constant voltage charging mode. The charging voltage is dynamically adjusted according to the battery characteristics and temperature. The adjustment formula is: Where V 标准 is the standard charging voltage, V max is the maximum allowable charging voltage, T max This is the maximum temperature allowed for the battery.

7. A digital sensor system according to claim 1, characterized in that: The wireless communication module of the aggregation unit adopts an adaptive coding and modulation algorithm based on signal quality assessment. This algorithm selects the coding and modulation method according to the signal-to-noise ratio SNR, bit error rate BER and channel fading situation of the received signal. The selection basis of the coding and modulation method is as follows: when SNR≥SNR1 and BER≤BER1, a high-order modulation and low-redundancy coding method is selected to improve the communication rate; when SNR2≤SNR<SNR1 or BER1<BER≤BER2, a moderate modulation and coding method is selected to balance the communication rate and reliability; when SNR<SNR2 or BER>BER2, a low-order modulation and high-redundancy coding method is selected to ensure the communication reliability, where SNR1, SNR2, BER1 and BER2 are thresholds preset according to system requirements.

8. A digital sensor system according to claim 1, characterized in that: The clamping structure of the collection unit adopts an intelligent adaptive clamping mechanism, which monitors the clamping force in real time through a built-in pressure sensor. The clamping force is adjusted based on the material properties, wire diameter and temperature of the wire. 抱紧 The calculation formula is: Where F0 is the initial clamping force, D is the current wire diameter, D0 is the standard wire diameter, E t is the elastic modulus of the current wire material, E0 is the elastic modulus of the standard wire material, k is the temperature coefficient, T is the current wire temperature, T0 is the standard temperature, and the clamping force is adjusted in real time.

9. A digital sensor system according to claim 1, characterized in that: The fault location algorithm of the master station adopts an intelligent location method based on machine learning. The neural network model is trained by using historical fault data and real-time monitoring data. The inputs of the model include line current, voltage, power factor, and fault indicator status information, and the output is the predicted value of the fault location. The structure and parameters of the neural network model are optimized by the backpropagation algorithm, and the connection weights between neurons are adjusted according to the prediction error. During the training process, the cross-validation method is adopted to improve the generalization ability of the model.

10. A digital sensor system according to claim 1, characterized in that: The wireless communication between the collection unit and the collection unit adopts a dynamic key negotiation encryption protocol. Before each communication, both parties generate a temporary session key based on a pre-shared key seed and a random number. The key generation formula is: K 会话 =Hash(K 种子 ||R3||R4), where Hash is the hash function, K 种子 R3 and R4 are random numbers generated by both parties for the shared key seed. By dynamically generating session keys, the security of communication is improved to prevent information leakage caused by key cracking.

Images