Remote monitoring and alarm system for power terminal based on internet of things technology

Abstract

The application belongs to the field of power terminal supervision, and is specifically a power terminal remote monitoring alarm system based on Internet of Things technology. Multidimensional operation data of the power terminal is collected by the Internet of Things, a terminal health index is calculated by a health degree evaluation module in combination with historical data, an operation trend is fitted according to time series data, an abnormality approximation coefficient is quantified, and a self-adaptive alarm threshold is dynamically generated with the health index and the abnormality coefficient as double factors, and a three-level progressive alarm is executed according to rules, so as to overcome the defects of traditional single-parameter and fixed-threshold determination, improve the state evaluation accuracy, predict the fault risk in advance, reduce false positives and false negatives, and reasonably allocate operation and maintenance resources through the hierarchical response mode, thereby greatly improving the intelligent level and fault disposal efficiency of the power terminal remote monitoring.

Description

Technical Field

[0001] This invention relates to the field of power terminal monitoring, specifically a remote monitoring and alarm system for power terminals based on Internet of Things (IoT) technology. Background Technology

[0002] As the core execution and sensing unit at the end of the distribution network, power terminals are widely distributed in various distribution nodes of urban residential communities, commercial complexes, industrial parks and remote rural power grids. They undertake key functions such as data acquisition, switch control and fault reporting. Their operational stability directly determines the continuity, reliability and power quality of regional power supply.

[0003] With the comprehensive advancement of the construction of new power systems, new loads such as distributed photovoltaics, energy storage devices, and electric vehicle charging piles are being connected to the distribution network on a large scale. This has resulted in the distribution network operating conditions exhibiting complex characteristics of high dynamism, large fluctuations, and numerous influencing factors, placing unprecedentedly high demands on the real-time status monitoring, early warning of anomalies, and operation and maintenance response efficiency of power terminals.

[0004] Currently, IoT communication technology has been fully integrated into the power terminal monitoring system, enabling real-time acquisition and remote transmission of various basic operating parameters such as current, voltage, internal temperature, and communication latency. However, conventional remote monitoring and alarm systems generally adopt a unified and fixed single threshold judgment mode, failing to combine the individual operating characteristics of different power terminals with the coupling relationship between multi-dimensional parameters to build a dynamic evaluation mechanism. Furthermore, they cannot predict the evolution trend of equipment health through time-series data analysis, resulting in alarm trigger sensitivity that cannot adapt to the actual needs of different operating stages of the terminal. At the same time, the single alarm response method is difficult to match different levels of abnormal risks, making it impossible to achieve accurate and efficient allocation of operation and maintenance resources.

[0005] Therefore, developing a graded alarm remote monitoring system for power terminals with dynamic evaluation of equipment health and trend prediction has become an urgent technical problem to be solved in the field of power terminal supervision. Summary of the Invention

[0006] The purpose of this invention is to provide a remote monitoring and alarm system for power terminals based on Internet of Things (IoT) technology, in order to solve the aforementioned technical deficiencies.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a remote monitoring and alarm system for power terminals based on Internet of Things (IoT) technology, including a health evaluation module, a status prediction engine module, a two-factor threshold mapping module, and a hierarchical alarm strategy module. It overcomes the technical defects of traditional single-threshold static alarms, ignoring parameter coupling relationships, fixed alarm sensitivity, and mismatch between response strength and risk, and realizes the intelligent upgrade of power terminal operating status from passive alarm to active prediction and from fixed threshold to dynamic adaptation.

[0008] The health assessment module calculates a normalized health index to quantify the degree of deviation of terminal operation based on the dispersion of multi-dimensional measured data and historical stable values; the state prediction engine module relies on the health time series within a continuous time window to fit the trend slope and solve the anomaly approximation coefficient to predict the urgency of the risk of the power terminal entering the abnormal range.

[0009] The dual-factor threshold mapping module uses the health index and the anomaly approximation coefficient as dual inputs to dynamically generate a dual-level alarm threshold that adapts to the real-time status of the power terminal; the hierarchical alarm strategy module executes a differentiated three-level alarm mode and matches the corresponding operation and maintenance response mechanism based on the dual-level alarm threshold and the terminal status parameters.

[0010] Furthermore, the health assessment module retrieves five types of parameters: effective current value, effective voltage value, internal ambient temperature, average telemetry message response delay, and number of telemetry change messages per unit time. It then calculates the health index by combining the mean and standard deviation of the stable operating data over the past 72 hours.

[0011] Furthermore, the state prediction engine module uses a fixed 30-second sampling interval to cache the health index of four consecutive sampling moments to form a four-dimensional time-series vector, and obtains the slope of the health trend by fitting it using the linear least squares method.

[0012] Furthermore, the status prediction engine module uses the health warning trigger threshold and alarm trigger threshold as benchmarks, and combines the trend slope to calculate the anomaly proximity coefficient, which quantitatively represents the urgency of the terminal being close to an abnormal state.

[0013] Furthermore, the dual-factor threshold mapping module pre-calibrates the first-level benchmark threshold and the dual-level benchmark interval value based on the power terminal type and rated power level, and dynamically corrects and generates two-level alarm thresholds by combining real-time health status and anomaly approximation coefficient.

[0014] Furthermore, the tiered alarm strategy module maintains normal monitoring when the anomaly approximation coefficient is lower than the first-level alarm threshold; and triggers a first-level alarm when the anomaly approximation coefficient is greater than or equal to the first-level alarm threshold but less than the second-level alarm threshold.

[0015] Furthermore, the tiered alarm strategy module triggers a second-level alarm when the anomaly approximation coefficient is greater than or equal to the second-level alarm threshold and the first-level alarm has not lasted for more than sixty seconds.

[0016] Furthermore, the tiered alarm strategy module directly triggers a tiered alarm when a tiered alarm lasts for more than 60 seconds, or when the health index is below 0.30 and the anomaly proximity coefficient is greater than or equal to the tiered alarm threshold.

[0017] Compared with the prior art, the beneficial effects of the present invention are:

[0018] 1. In this invention, by combining multi-dimensional operating parameters with historical stable data to calculate the health of power terminals, the accuracy of equipment status assessment is effectively improved. Furthermore, by analyzing time-series data to fit operating trends and quantifying anomaly approximation coefficients, fault risks can be predicted in advance. At the same time, a two-factor dynamic threshold mechanism is adopted to adaptively adjust alarm sensitivity, which greatly reduces false alarms caused by operating condition disturbances and missed alarms caused by hidden faults.

[0019] 2. In this invention, by setting a three-level progressive alarm strategy, the alarm response intensity is matched with the urgency of the equipment abnormality. Differentiated prompt information is pushed for different levels of alarms, and maintenance resources are reasonably allocated to shorten the fault handling time and improve the intelligence level of remote monitoring of power terminals and the overall maintenance efficiency. Attached Figure Description

[0020] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings;

[0021] Figure 1 This is a schematic diagram of the overall system structure of the present invention;

[0022] Figure 2 This is a schematic diagram of the operation method steps in this invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] Reference Figure 1-2 As shown, the power terminal remote monitoring and alarm system based on Internet of Things (IoT) technology proposed in this invention directly calls the operating parameters of the power terminal (which has conventional electrical parameter sensing and acquisition capabilities, and the acquired data is uploaded to the monitoring platform in real time through the IoT communication channel) and sends them to the monitoring platform via the IoT communication network. The system consists of a health evaluation module, a status prediction engine module, a two-factor threshold mapping module, and a hierarchical alarm strategy module. The modules transmit processing results through an internal data bus, forming a complete closed-loop control link from terminal status perception to alarm strategy output.

[0025] The health assessment module constructs a normalized evaluation index for the health status of power terminals based on the dispersion between measured values ​​and historical stable values ​​of multi-dimensional operating parameters. This quantifies the deviation of each terminal from its historical optimal operating range, overcoming the shortcomings of traditional single-threshold judgment methods that ignore the coupling relationship between parameters and lack a basis for fixed weights. This significantly improves the accuracy of the health assessment results. The specific operation process is as follows:

[0026] The health assessment module retrieves the effective current value I, effective voltage value U, internal ambient temperature T, average telemetry message response delay D, and the number of telemetry change messages F per unit time from the real-time database of the monitoring platform. Simultaneously, it extracts the recorded values ​​of these parameters from the historical database when the terminal was in a stable operating state over the past 72 hours. Based on the core analysis of the dispersion between the current measured values ​​and historical stable values, a formula for calculating the health index H is proposed:

[0027] ;

[0028] Where N represents the total number of parameter dimensions involved in the health assessment. Here, N is set to 5, and X1, X2, X3, X4, and X5 represent the effective value of current I, effective value of voltage U, internal ambient temperature T, average telemetry message response delay D, and the number of times remote signal change messages occur per unit time F, respectively.

[0029] M1 to M5 represent the average values ​​of the same parameter under historical stable operating conditions. The method of obtaining them is to take the arithmetic average of the recorded values ​​of each parameter that were in a stable operating state (i.e., no alarms at the terminal and voltage and current fluctuations not exceeding ±5% of the rated value) over the past 72 hours.

[0030] S1 to S5 represent the standard deviation of each parameter under historical stable operating conditions. The method of obtaining them is as follows: calculate the standard deviation of each parameter based on the same 72 hours of stable data. W1 to W5 represent the weight coefficients of each parameter, and the five weight coefficients satisfy W1+W2+W3+W4+W5=1.

[0031] It should be noted that the health index H ranges from [0,1]. The closer the H value is to 1, the better the operating status of the power terminal. The closer the H value is to 0, the more serious the deviation of the power terminal from the normal state. After the health evaluation module finishes running, it outputs the health index H value to the status prediction engine module.

[0032] The status prediction engine module takes the health index sequence within a continuous time window as input and predicts the probability of the terminal's operating status entering an abnormal range in the future by analyzing the temporal evolution trend of the health index. The specific operation process is as follows:

[0033] The state prediction engine module receives the current health index Ht and three sets of historical health indices Ht1, Ht2 and Ht3 cached in its internal cache from the health evaluation module. Ht1 is the health index at the previous sampling time, Ht2 is the health index at the two previous sampling times, and Ht3 is the health index at the three previous sampling times. Preferably, the sampling time interval is fixed at thirty seconds.

[0034] A four-dimensional health vector [Ht3, Ht2, Ht1, Ht] is constructed according to the time sequence. The trend slope K corresponding to this vector is calculated using the linear least squares fitting method, and the formula is as follows:

[0035] ;

[0036] Where j is the sampling point number, ranging from 1 to 4; Tj is the absolute timestamp corresponding to the j-th sampling point, which is directly read from the timestamp field of the monitoring platform;

[0037] Hj is the health index value of the j-th sampling point; Ta is the arithmetic mean of the timestamps of the four sampling points; Ha is the arithmetic mean of the four health index values.

[0038] After obtaining the trend slope K, the state prediction engine module further calculates the anomaly approximation coefficient P, using the following formula: ;

[0039] Wherein, Hu is the warning trigger threshold of the health index, preferably 0.65; Hd is the alarm trigger threshold of the health index, preferably 0.45; A is the trend influence intensity coefficient, preferably 0.35; Km is the upper limit of the absolute value of the slope of the health trend under normal conditions, preferably 0.15; the max function means taking the larger value between 0 and the expression in parentheses to ensure that the abnormal approximation coefficient P is a non-negative number.

[0040] It should be noted that the calculated result of the abnormal approach coefficient P can be greater than 1. A P value less than 1 indicates that the current health of the power terminal is still within a certain margin of the alarm threshold. A P value greater than or equal to 1 indicates that the health has dropped to near the alarm trigger threshold or the trend slope indicates that it will approach the threshold in a short period of time. The larger the P value, the higher the urgency of the risk of the power terminal entering an abnormal state. The state prediction engine module outputs the calculated abnormal approach coefficient P together with the current health index Ht to the two-factor threshold mapping module.

[0041] The dual-factor threshold mapping module uses the health index and the anomaly approximation coefficient as two input factors. Through nonlinear function mapping, it determines a two-level alarm threshold that adapts to the current terminal status. This enables power terminals with good health to implement a more lenient alarm strategy to avoid frequent false alarms caused by small disturbances, while power terminals with poor health implement a more sensitive alarm strategy to detect abnormal signals as early as possible. The specific operation process is as follows:

[0042] The two-factor threshold mapping module receives the health index Ht and the anomaly approximation coefficient P from the state prediction engine module. It sets the first-level alarm threshold R1 as a modified form of the modular baseline threshold, and the second-level alarm threshold R2 as the sum of the first-level alarm threshold and the difference between the two levels. The two-factor mapping formula for the first-level alarm threshold R1 and the difference between the two levels Rd is defined as follows:

[0043] ;

[0044] ;

[0045] ;

[0046] Rb is the first-level benchmark threshold. The specific value is determined in advance by the benchmark threshold table in the engineering manual according to the terminal type (such as feeder terminal, distribution transformer terminal, switch station terminal) and rated power level. For example, for a distribution transformer terminal with a rated power of 100 kilowatts, Rb is 0.60.

[0047] Rb2 is the baseline interval value between the two levels, which is also pre-calibrated according to the terminal type and power level. For example, for the same distribution transformer terminal, Rb2 is 0.25; B is the health correction coefficient, preferably 0.20; G is the risk aggravation correction coefficient, preferably 0.40.

[0048] Furthermore, in the mapping formula, (1-Ht) represents the degree to which the health deviates from the perfect state, and P / (1+P) is the compression transformation of the abnormal approximation coefficient, mapping its value range from [0,+∞) to [0,1).

[0049] It should be noted that this mapping process ensures that when the terminal health Ht is high (close to 1) and the anomaly approximation coefficient P is low (close to 0), R1 is close to the baseline value Rb, Rd is close to the baseline interval value Rb2, and the two-level alarm intervals are within the normal configuration range.

[0050] When Ht decreases or P increases, R1 is adjusted downwards in the more sensitive direction (the value decreases), while Rd is widened, giving the second-level alarm a wider response range; after the two-factor threshold mapping module finishes running, it outputs the alarm thresholds of the two levels, R1 and R2, to the hierarchical alarm strategy module.

[0051] The tiered alarm strategy module, based on the dual-level alarm thresholds output by the dual-factor threshold mapping module and the anomaly approximation coefficient and health index output by the state prediction engine module, executes a progressive three-level alarm mode switching and execution. Each alarm level corresponds to a different level of abnormal state (Level 1 alarms provide a light reminder to on-duty personnel, Level 2 alarms directly contact frontline maintenance personnel, and Level 3 alarms trigger a network-wide emergency response based on the duration of continuous deterioration or extremely low health values). This ensures that the alarm response strength is naturally matched to the urgency of the anomaly, significantly improving the response efficiency and resource utilization rationality of the operation and maintenance system. The specific operation process is as follows:

[0052] The hierarchical alarm strategy module obtains the first-level alarm threshold R1 and the second-level alarm threshold R2 from the two-factor threshold mapping module, continuously compares the abnormal approximation coefficient P output by the state prediction engine module in real time, and monitors the changes in the health index Ht.

[0053] When the abnormal approach coefficient P is continuously lower than the first-level alarm threshold R1, the power terminal is determined to be in the normal range and no alarm action is triggered. Only the background recording and updating of the health index is maintained. When the abnormal approach coefficient P rises to greater than or equal to R1 but less than R2, it is determined to enter the first-level alarm mode. At this time, the alarm device on the edge side emits low-frequency light flashing and low volume warning on the spot. At the same time, the alarm summary information is pushed to the primary prompt window of the operation and maintenance duty platform, including the terminal identifier, health index Ht, abnormal approach coefficient P and trigger time.

[0054] When the anomaly approximation coefficient P rises further to greater than or equal to R2, the system first determines whether the current level 1 alarm mode has been maintained for more than 60 seconds. If it has not exceeded 60 seconds, it enters the level 2 alarm mode. If it has been maintained for more than 60 seconds, it directly enters the level 3 alarm mode. In addition, when the health index Ht is lower than 0.30 and the anomaly approximation coefficient P is greater than or equal to R1, the system also directly activates the level 3 alarm mode to deal with special scenarios where the health has been severely deteriorated but the anomaly approximation coefficient has not yet responded sufficiently.

[0055] Furthermore, the execution method of the secondary alarm mode is as follows: the alarm device on the edge side is converted to high-frequency flashing light and loud continuous alarm on the spot. At the same time, it pushes emergency alarm information with location information and predicted fault type to the mobile terminal of the associated inspection personnel through the Internet of Things communication channel, and sends a pop-up forced reminder to the centralized control interface of the equipment management center.

[0056] The execution method of the Level 3 alarm mode is as follows: broadcast alarm commands are issued to all associated operation and maintenance terminals, and the geographical location and topology node of the power terminal are locked and highlighted on the monitoring screen to form a visual alarm situation map, and the work order system is automatically triggered to generate the highest priority emergency repair task.

[0057] The working principle of this invention is as follows: During use, it relies on the Internet of Things to collect multi-dimensional operational data from power terminals, including electrical parameters, internal temperature, and communication latency. Combined with historical parameter dispersion, a normalized health index is calculated to improve the accuracy of terminal health status assessment. Subsequently, based on continuous time-series health data, a deterioration trend is fitted, and an anomaly approximation coefficient is quantified to enable early prediction of fault risks. Then, a two-level adaptive alarm threshold is dynamically generated using both health and anomaly coefficients. The alarm sensitivity is flexibly adjusted according to the real-time status of the power terminal, effectively reducing false alarms caused by small disturbances and missed alarms caused by hidden faults. Finally, a three-level progressive alarm strategy is matched, ensuring that the response intensity accurately corresponds to the risk level, optimizing the allocation of operation and maintenance resources, and significantly improving the intelligence level of remote monitoring of power terminals and the efficiency of fault handling.

[0058] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, enabling those skilled in the art to better understand and utilize it. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A remote monitoring and alarm system for power terminals based on Internet of Things (IoT) technology, characterized in that: It includes a health assessment module, a status prediction engine module, a two-factor threshold mapping module, and a graded alarm strategy module; The health assessment module calculates a normalized health index to quantify the degree of deviation in terminal operation based on the dispersion of multi-dimensional measured data and historical stable values. The status prediction engine module relies on the health time series within a continuous time window to fit the trend slope and solve the anomaly approximation coefficient to predict the urgency of the risk of the power terminal entering the abnormal range. The dual-factor threshold mapping module uses the health index and the anomaly approximation coefficient as dual inputs to dynamically generate a dual-level alarm threshold that adapts to the real-time status of the power terminal. The hierarchical alarm strategy module executes differentiated three-level alarm modes and matches the corresponding operation and maintenance response mechanisms based on the two-level alarm thresholds and terminal status parameters.

2. The power terminal remote monitoring and alarm system based on Internet of Things technology according to claim 1, characterized in that, The health assessment module retrieves five types of parameters: effective current value, effective voltage value, internal ambient temperature, average telemetry message response delay, and number of remote signaling change messages per unit time. It then calculates the health index by combining the mean and standard deviation of the stable operation data over the past 72 hours.

3. The power terminal remote monitoring and alarm system based on Internet of Things technology according to claim 1, characterized in that, The state prediction engine module uses a fixed 30-second sampling interval and caches the health index of four consecutive sampling times to form a four-dimensional time-series vector. The slope of the health trend is obtained by fitting the vector using the linear least squares method.

4. The power terminal remote monitoring and alarm system based on Internet of Things technology according to claim 3, characterized in that, The status prediction engine module uses the health warning trigger threshold and alarm trigger threshold as benchmarks, and calculates the anomaly proximity coefficient by combining the trend slope to quantitatively characterize the urgency of the terminal being close to an abnormal state.

5. The power terminal remote monitoring and alarm system based on Internet of Things technology according to claim 1, characterized in that, The dual-factor threshold mapping module pre-calibrates the first-level benchmark threshold and the dual-level benchmark interval value based on the power terminal type and rated power level, and dynamically corrects and generates two-level alarm thresholds by combining real-time health status and anomaly approximation coefficient.

6. The power terminal remote monitoring and alarm system based on Internet of Things technology according to claim 1, characterized in that, The tiered alarm strategy module maintains normal monitoring when the anomaly approximation coefficient is lower than the first-level alarm threshold; and triggers a first-level alarm when the anomaly approximation coefficient is greater than or equal to the first-level alarm threshold but less than the second-level alarm threshold.

7. The power terminal remote monitoring and alarm system based on Internet of Things technology according to claim 6, characterized in that, The tiered alarm strategy module triggers a second-level alarm when the anomaly approximation coefficient is greater than or equal to the second-level alarm threshold and the first-level alarm has not lasted for more than 60 seconds.

8. The power terminal remote monitoring and alarm system based on Internet of Things technology according to claim 7, characterized in that, The tiered alarm strategy module will directly trigger a tiered alarm if a tiered alarm lasts for more than 60 seconds, or if the health index is below 0.30 and the anomaly proximity coefficient is greater than or equal to the tiered alarm threshold.

Images