Steel market information big data remote analysis management platform

Abstract

The invention discloses a steel market big data remote analysis management platform, which comprises a multi-source data acquisition module, a data extraction and analysis module, an abnormal data processing module and a storage module, and is characterized in that the abnormal data processing module comprises a signal conditioning circuit, a DSP and a remote communication module. The signal conditioning circuit is designed to carry out composite modulation and optimization on abnormal data signals, identification of abnormal values and fine processing of the abnormal data signals are combined, the accuracy and depth of data analysis can be remarkably improved by means of the DSP technology, and a powerful guarantee is provided for accurate identification of abnormal features. The steel market is predicted and analyzed through the isolated forest model, the prediction accuracy can be effectively improved, abnormal values in the data can be quickly and accurately identified and processed, and powerful support is provided for decision making of the steel industry.

Description

Technical Field

[0001] The present invention relates to the technical field of steel market analysis, and in particular to a remote analysis and management platform for steel market big data. Background Art

[0002] With the rapid development of global industrialization and urbanization, steel, as a basic material, plays an important role in many fields such as construction, automobiles, and machinery manufacturing. The fluctuation of the steel market has a profound impact on the cost control, production planning, and even the stable development of the entire economy of related industries. Therefore, accurate and timely grasp of steel market data is of great significance for enterprises and government departments to formulate strategies and optimize resource allocation.

[0003] The data sources of steel enterprises are extensive and diverse, including multiple information systems such as ERP system, MES system, measurement and inspection system, etc. Due to the diversity of data sources, there is a lack of unified standards and specifications in the process of data collection, processing and analysis, resulting in uneven data quality and difficulty in effective analysis and utilization. In response to this technical problem, the existing patent document with publication number CN116662295 A proposes a method for building a unified data model for the steel industry based on a big data center. This method combs the business through the two dimensions of entity business objects and business processes, constructs a unified subject domain model for the industry, a standard tree of entity business objects and a standard tree of business processes, and uses knowledge graph technology to construct an industry concept graph model. However, this method has limitations in the dynamic update and maintenance of the data model, and it is difficult to adapt to the rapidly changing market environment; and it has limited capabilities in the detection and processing of abnormal data, and cannot detect and respond to market anomalies in a timely manner.

[0004] In addition, Tempodata announced a project construction plan for a big data analysis service platform for the steel industry. The plan aims to achieve scientific decision-making through data mining technology, including big data analysis in production, quality, energy and other aspects. The platform focuses on the data processing and analysis capabilities of the background. Although data mining technology has improved the analysis efficiency to a certain extent, in the face of massive steel market data, a single data mining method is difficult to accurately identify all outliers. Moreover, outliers, as "noise" in the data, may contain important market information, but may also interfere with and mislead the data analysis results. Therefore, after identifying outliers, how to conduct effective and refined processing has become an urgent problem to be solved.

[0005] In view of the above situation, the present invention aims to provide a new solution to solve this problem. Summary of the invention

[0006] In view of the above situation, in order to overcome the defects of the prior art, the purpose of the present invention is to provide a remote analysis and management platform for steel market big data.

[0007] The technical solution is: a remote analysis and management platform for steel market big data, including:

[0008] Multi-source data collection module, used to collect raw data related to steel market conditions from multiple channels in real time or at regular intervals;

[0009] The data extraction and analysis module pre-processes the collected raw data and extracts key information, and uses big data analysis models to identify and analyze outliers in the data;

[0010] An abnormal data processing module is used to receive the abnormal data signal from the data extraction and analysis module, and transmit the signal to the user management platform after signal conditioning;

[0011] The storage module is connected to the data extraction and analysis module and is used to store the collected data and analysis results.

[0012] Preferably, the abnormal data processing module includes:

[0013] A signal conditioning circuit, used for performing complex modulation on the abnormal data signal to optimize signal characteristics;

[0014] DSP, used to analyze and process the conditioned signal to identify abnormal features in the data;

[0015] The remote communication module is used to remotely transmit the abnormal data processed by the DSP to the user management platform.

[0016] Preferably, the signal conditioning circuit comprises:

[0017] A signal optimization unit, configured to perform limiting amplification on the abnormal data signal and apply a reference voltage during the amplification process to optimize the dynamic range of the abnormal data signal;

[0018] A waveform modulation unit, used for converting the low-frequency signal after limited amplification into a high-frequency signal through frequency modulation;

[0019] A denoising unit, used to remove noise and interference components from the high-frequency signal to improve the signal-to-noise ratio of the signal;

[0020] The sampling feedback unit is used to sample the denoised signal and implement closed-loop feedback regulation processing on the signal conditioning circuit.

[0021] Preferably, the signal optimization unit includes an op amp AR1, an op amp AR2 and an adjustable power supply component, the inverting input terminal of the op amp AR1 is connected to resistor R1, resistor R2, resistor R3, one end of capacitor C1, the cathode of diode VD1 and the anode of diode VD2, the other end of resistor R1 and capacitor C1 is connected to the abnormal data signal input port, the other end of resistor R2 is grounded, the in-phase input terminal of the op amp AR1 is connected to the adjustable power supply component, the output terminal of the op amp AR1 is connected to the anode of diode VD1, the cathode of diode VD2 and the in-phase input terminal of the op amp AR2, and the inverting input terminal of the op amp AR2 is connected to the other end of resistor R3, the output terminal of the op amp AR2 and the denoising unit.

[0022] Preferably, the adjustable power supply component includes a voltage zener diode DZ1, an adjustable resistor VR1 and an op amp AR3, the anode of the voltage zener diode DZ1 and one end of the adjustable resistor VR1 are connected to a +12V power supply through a resistor R4, the cathode of the voltage zener diode DZ1 and the other end of the adjustable resistor VR1 are grounded, the adjustment end of the adjustable resistor VR1 is connected to the in-phase input end of the op amp AR3 and one end of the resistor R5, the inverting input end of the op amp AR3 is grounded through a resistor R7, the output end of the op amp AR3 is connected to the other end of the resistor R5, and is connected to the in-phase input end of the op amp AR1 through a resistor R6.

[0023] Preferably, the waveform modulation unit includes a JFET tube Q1 and a square wave generator, and the JFET tube Q1 is arranged between the operational amplifier AR1 and the operational amplifier AR2; the drain of the JFET tube Q1 is connected to the output end of the operational amplifier AR1, the source of the JFET tube Q1 is connected to the in-phase input end of the operational amplifier AR2, and the gate of the JFET tube Q1 is connected to the signal output end of the square wave generator.

[0024] Preferably, the square wave generator includes an op amp AR4, an inverting input end of the op amp AR4 is connected to one end of resistors R8 and R9, and is grounded through a capacitor C2, the other end of the resistor R8 is connected to the anode of a diode VD3, the cathode of the diode VD3 is connected to one end of an adjustable resistor VR2, the other end of the resistor R9 is connected to the cathode of a diode VD4, the anode of the diode VD4 is connected to the other end of the adjustable resistor VR2, a non-inverting input end of the op amp AR4 is connected to one end of a resistor R10 and R11, the other end of the resistor R10 is grounded, an output end of the op amp AR4 is connected to the other end of the resistor R11, one end of a bidirectional voltage regulator D1, an adjustment end of the adjustable resistor VR2 and a gate of a MOS tube Q1 through a resistor R12, and the other end of the bidirectional voltage regulator D1 is grounded.

[0025] Preferably, the denoising unit includes a resistor R13, a capacitor C3, a capacitor C4 and an inductor L1, one end of the resistor R13, the capacitor C3 and the inductor L1 are connected to the output end of the op amp AR2, the other end of the resistor R13 and the capacitor C3 are grounded, and the other end of the inductor L1 is connected to the DSP and the sampling feedback unit, and is grounded through the capacitor C4.

[0026] Preferably, the sampling feedback unit includes a resistor R14 and a resistor R15, one end of the resistor R14 is connected to the output end of the denoising unit, the other end of the resistor R14 is connected to the inverting input end of the operational amplifier AR3, and is grounded through the resistor R15.

[0027] Preferably, the data extraction and analysis module uses an isolation forest model to identify outliers in the data.

[0028] Through the above technical solution, the beneficial effects of the present invention are:

[0029] 1. The present application designs a signal conditioning circuit to perform complex modulation and optimization on abnormal data signals. The signal conditioning circuit includes multiple components such as a signal optimization unit, a waveform modulation unit, a denoising unit, and a sampling feedback unit. Through steps such as limiting amplification, frequency modulation, denoising processing, and closed-loop feedback adjustment, the signal's anti-interference ability and transmission efficiency are significantly improved;

[0030] 2. This application combines the identification of outliers with the refined processing of abnormal data signals, and with the help of DSP technology, can significantly improve the accuracy and depth of data analysis, providing a strong guarantee for the accurate identification of abnormal features;

[0031] 3. This application uses the isolation forest model to predict and analyze the steel market, which can not only effectively improve the accuracy of the prediction, but also quickly and accurately identify and process outliers in the data, providing strong support for decision-making in the steel industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a structural block diagram of the system module of the present invention.

[0033] Figure 2 It is a structural block diagram of the abnormal data processing module in the present invention.

[0034] Figure 3 It is a circuit schematic diagram of the signal optimization unit in the present invention.

[0035] Figure 4 It is the overall principle diagram of the signal conditioning circuit in the present invention.

[0036] Figure 5 The present invention is a flow chart of the method for detecting outliers in steel market data based on the isolation forest model. DETAILED DESCRIPTION

[0037] The above and other technical contents, features and effects of the present invention are described in detail below with reference to the attached Figure 1 To Attachment Figure 5 The detailed description of the embodiments will clearly show that the structural contents mentioned in the following embodiments are all based on the drawings in the specification.

[0038] Exemplary embodiments of the present invention will be described below with reference to the accompanying drawings.

[0039] like Figure 1 As shown, a remote analysis and management platform for steel market big data includes:

[0040] Multi-source data collection module, used to collect raw data related to steel market conditions from multiple channels in real time or at regular intervals;

[0041] The data extraction and analysis module pre-processes the collected raw data and extracts key information, and uses big data analysis models to identify and analyze outliers in the data;

[0042] The abnormal data processing module is used to receive the abnormal data signal from the data extraction and analysis module, and transmit it to the user management platform after signal conditioning;

[0043] The storage module is connected to the data extraction and analysis module and is used to store the collected data and analysis results.

[0044] In the specific implementation process, the multi-source data acquisition module can use a variety of technical means such as web crawlers, API interface calls, and IoT sensors to automatically or regularly collect key raw data of the steel market from multiple data sources, such as prices, inventory, production, policies, etc. The data extraction and analysis module cleans, integrates, and extracts key information from the collected raw data, and uses big data analysis models for in-depth mining to automatically identify and analyze outliers in the data. When the model identifies an outlier, it will generate a corresponding abnormal data signal and send it to the abnormal data processing module for refined processing.

[0045] Specifically, Figure 2 As shown, the abnormal data processing module includes:

[0046] A signal conditioning circuit, used for performing complex modulation on abnormal data signals to optimize signal characteristics;

[0047] DSP (digital signal processor), used to analyze and process the conditioned signal to identify abnormal features in the data;

[0048] The remote communication module is used to remotely transmit the abnormal data processed by DSP to the user management platform.

[0049] Among them, the signal conditioning circuit includes:

[0050] A signal optimization unit, used for limiting and amplifying the abnormal data signal, and applying a reference voltage during the amplification process to optimize the dynamic range of the abnormal data signal;

[0051] A waveform modulation unit, used for converting the low-frequency signal after limited amplification into a high-frequency signal through frequency modulation;

[0052] De-noising unit, used to remove noise and interference components from high-frequency signals and improve the signal-to-noise ratio;

[0053] The sampling feedback unit is used to sample the denoised signal and implement closed-loop feedback regulation processing on the signal conditioning circuit.

[0054] In the specific implementation process, the signal conditioning circuit performs a series of processing and optimization on the abnormal data signal to ensure that the signal can maintain high quality and accuracy in the subsequent digital processing process. The following is a detailed description of the components of the signal conditioning circuit:

[0055] The signal optimization unit first receives the original abnormal data signal and optimizes its dynamic range. Specifically, Figure 3 As shown, the signal optimization unit includes an op amp AR1, an op amp AR2 and an adjustable power supply component. The inverting input terminal of the op amp AR1 is connected to a resistor R1, a resistor R2, a resistor R3, one end of a capacitor C1, a cathode of a diode VD1 and an anode of a diode VD2. The other ends of the resistor R1 and the capacitor C1 are connected to an abnormal data signal input port. The other end of the resistor R2 is grounded. The in-phase input terminal of the op amp AR1 is connected to the adjustable power supply component. The output terminal of the op amp AR1 is connected to the anode of the diode VD1, the cathode of the diode VD2 and the in-phase input terminal of the op amp AR2. The inverting input terminal of the op amp AR2 is connected to the other end of the resistor R3, the output terminal of the op amp AR2 and the denoising unit.

[0056] In the above, the original abnormal data signal is input to the signal optimization unit through the input port INT, and first passes through the RC filter network composed of resistor R1, resistor R2 and capacitor C1 to preliminarily filter out the high-frequency noise in the signal to smooth the signal. The operational amplifier AR1 is used as the first-stage amplifier to enhance the amplitude of the abnormal data signal. The diodes VD1 and VD2 play the role of clamping and limiting in the amplification process of the operational amplifier AR1 to limit the maximum and minimum amplitudes of the output signal to prevent the signal from exceeding the processing range of the subsequent DSP and avoid signal distortion. The operational amplifier AR2 is used as the second-stage amplifier, and its inverting input terminal is connected to the output terminal to form negative feedback, which is used to stabilize the amplification factor and further reduce nonlinear distortion.

[0057] At the same time, the adjustable power supply component applies a reference voltage during the amplification process to ensure that the amplifier circuit can maintain consistent gain characteristics under different working conditions. Figure 4 As shown, the adjustable power supply component includes a voltage zener diode DZ1, an adjustable resistor VR1 and an op amp AR3, the anode of the voltage zener diode DZ1 and one end of the adjustable resistor VR1 are connected to a +12V power supply through a resistor R4, the cathode of the voltage zener diode DZ1 and the other end of the adjustable resistor VR1 are grounded, the adjustment end of the adjustable resistor VR1 is connected to the in-phase input end of the op amp AR3 and one end of the resistor R5, the inverting input end of the op amp AR3 is grounded through a resistor R7, the output end of the op amp AR3 is connected to the other end of the resistor R5, and is connected to the in-phase input end of the op amp AR1 through a resistor R6.

[0058] Among them, +12V power supply, resistor R4, voltage regulator diode DZ1 and adjustable resistor VR1 constitute a voltage regulator. The voltage regulator diode DZ1 uses its reverse breakdown characteristics to play a voltage stabilizing role in the circuit and provide a relatively stable voltage reference for the subsequent circuit; at the same time, adjusting the resistance value of VR1 can adjust the output voltage of the voltage regulator. Finally, the voltage follower function of the op amp AR3 outputs a stable reference voltage to the op amp AR1. The advantage of the adjustable power supply component applying a reference voltage during the amplification process is that it can provide a stable and precisely adjustable voltage reference, ensuring that the amplifier circuit can maintain consistent gain characteristics under different working conditions, thereby improving the accuracy and reliability of signal amplification, while enhancing the flexibility and adaptability of the system.

[0059] In order to further improve the anti-interference ability and transmission efficiency of abnormal data signals, a waveform modulation unit is used to convert the low-frequency signal after limiting amplification into a high-frequency signal through frequency modulation. Figure 4 As shown, the waveform modulation unit includes a JFET tube Q1 and a square wave generator. The JFET tube Q1 is arranged between the operational amplifier AR1 and the operational amplifier AR2; the drain of the JFET tube Q1 is connected to the output end of the operational amplifier AR1, the source of the JFET tube Q1 is connected to the in-phase input end of the operational amplifier AR2, and the gate of the JFET tube Q1 is connected to the signal output end of the square wave generator.

[0060] Wherein, the square wave generator includes an operational amplifier AR4, an inverting input end of the operational amplifier AR4 is connected to one end of the resistor R8 and the resistor R9, and is grounded through a capacitor C2, the other end of the resistor R8 is connected to the anode of the diode VD3, the cathode of the diode VD3 is connected to one end of the adjustable resistor VR2, the other end of the resistor R9 is connected to the cathode of the diode VD4, the anode of the diode VD4 is connected to the other end of the adjustable resistor VR2, the non-inverting input end of the operational amplifier AR4 is connected to one end of the resistor R10 and the resistor R11, the other end of the resistor R10 is grounded, the output end of the operational amplifier AR4 is connected to the other end of the resistor R11, one end of the bidirectional voltage regulator D1, the adjustment end of the adjustable resistor VR2 and the gate of the MOS tube Q1 through a resistor R12, and the other end of the bidirectional voltage regulator D1 is grounded.

[0061] In the waveform modulation process, the inverting input terminal of the operational amplifier AR4 forms an RC oscillation circuit through resistors R8, R9 and capacitor C2. The square wave generator can change the parameters of the oscillation circuit by adjusting the resistance value of the adjustable resistor VR2, thereby adjusting the frequency of the output square wave. The output square wave signal is sent to the gate of the JFET tube Q1 as a modulation signal, and the JFET tube Q1 is set between the operational amplifier AR1 and the operational amplifier AR2, which is used to modulate the abnormal data signal on the signal transmission path, that is, the square wave signal controls the conductivity between the drain and the source of the JFET tube Q1 by changing the gate voltage of the JFET tube Q1, thereby realizing the frequency modulation of the low-frequency signal. The modulated abnormal data signal has good anti-interference performance, and its high-frequency signal is not easily affected by low-frequency noise and interference during the transmission process, thereby improving the identifiability of abnormal data.

[0062] Furthermore, the denoising unit removes noise and interference components from the modulated abnormal data signal. Specifically, Figure 4 As shown, the denoising unit includes a resistor R13, a capacitor C3, a capacitor C4 and an inductor L1, one end of the resistor R13, the capacitor C3 and the inductor L1 are connected to the output end of the operational amplifier AR2, the other end of the resistor R13 and the capacitor C3 are grounded, and the other end of the inductor L1 is connected to the DSP and the sampling feedback unit and grounded through the capacitor C4. Due to the influence of external factors during signal transmission, abnormal data signals may appear as irregular forms such as sudden changes, spikes, and jitters of the signal. Therefore, the RC-LC denoising network is used to effectively remove noise and interference components from the high-frequency signal, thereby improving the signal-to-noise ratio and quality of the signal.

[0063] At the same time, in order to enhance the stability of the abnormal data signal processing system, a sampling feedback unit is introduced to monitor and adjust the signal quality in real time. Specifically, the sampling feedback unit includes a resistor R14 and a resistor R15, one end of the resistor R14 is connected to the output end of the denoising unit, and the other end of the resistor R14 is connected to the inverting input end of the op amp AR3, and is grounded through the resistor R15. Among them, the sampling feedback unit samples the denoised signal using the resistor shunt principle, and the adopted signal is fed back to the inverting input end of the op amp AR3 in the adjustable power supply component, and the op amp AR3 automatically adjusts the output of its reference voltage using the subtractor principle. Through the collaborative work of the sampling feedback unit and the op amp AR3, the system can sense and respond to abnormal fluctuations in real time. When abnormal fluctuations occur in the system, the sampling signal will change. In order to maintain the stability of the system, AR3 will automatically adjust the output of its reference voltage to compensate for this change, thereby suppressing the fluctuation and restoring the stable state of the system. This mechanism greatly improves the robustness and reliability of the system, enabling it to operate stably in various complex environments.

[0064] By performing composite modulation on the abnormal data signal through the above-mentioned signal conditioning circuit, the characteristics of the signal are optimized, making it more suitable for subsequent digital processing. Next, the DSP is responsible for further analyzing and processing the conditioned signal, extracting and analyzing the characteristic parameters of the abnormal data signal to accurately identify the abnormal features in the data. After identifying the abnormal features, the DSP also needs to make intelligent judgments based on business logic, which includes combining the abnormal features with the system's operating status, fault history, environmental conditions, etc., to evaluate the possible causes and impacts of the abnormality. Finally, the remote communication module remotely transmits the abnormal data processed by the DSP and its analysis results to the user management platform for real-time monitoring and management.

[0065] As another embodiment of the present application, in order to improve the accuracy of predictive analysis, it is particularly important to timely and accurately identify outliers in steel market data. In this scenario, the data extraction and analysis module uses the isolation forest model to identify outliers in the data. Isolation Forest is an anomaly detection algorithm based on ensemble learning, which uses a tree structure to isolate abnormal points. Unlike most density-based anomaly detection algorithms, Isolation Forest assumes that outliers are few and different, and they are more likely to be "isolated" (that is, separated from most data points). The algorithm constructs multiple isolation trees by randomly selecting features, and calculates the average path length from each data point to its nearest root node to evaluate the degree of abnormality of the data. Figure 5 As shown, the following are the specific method steps for detecting outliers in steel market data based on the isolation forest model:

[0066] First, the collected steel market data is cleaned and preprocessed, including removing noise, filling missing values, standardizing or normalizing features, etc., to ensure data quality;

[0067] According to business needs and model characteristics, select features that have a significant impact on steel market forecasts, such as price, transaction volume, inventory, raw material prices, etc.;

[0068] Input the preprocessed data into the isolation forest model, which will automatically build multiple isolation trees and calculate the anomaly score for each data point;

[0069] According to actual needs or data distribution, a threshold of anomaly score is set, and data points with anomaly scores lower than the threshold are marked as outliers.

[0070] In the above, the calculation process of the anomaly score is as follows:

[0071] For each data point, calculate the path length from the root node to the leaf node;

[0072] Since different trees may have different depths, the path length needs to be standardized. The formula C(n) = 2H(n-1)-(2(n-1) / n is used to estimate the expected value of the average path length H(n) given the number of samples n, where C(n) is the standardized path length.

[0073] The anomaly score of a data point is the average of its normalized path lengths in all trees. The lower the score, the more abnormal the data point. In the specific implementation, the specific algorithm is as follows:

[0074]

[0075]

[0076] In the above, the path_length function needs to calculate the path length based on the specific tree structure and data points, while the select_feature_and_split and split functions are responsible for randomly selecting features and split values ​​and splitting the data set into two parts.

[0077] Using the above-mentioned isolation forest model to predict and analyze steel market conditions can not only effectively improve the accuracy of predictions, but also quickly and accurately identify and process outliers in the data, providing strong support for decision-making in the steel industry.

[0078] In summary, in the abnormal data processing module, the present application designs a signal conditioning circuit to perform composite modulation and optimization on the abnormal data signal. The signal conditioning circuit includes multiple components such as a signal optimization unit, a waveform modulation unit, a denoising unit, and a sampling feedback unit. Through steps such as limiting amplification, frequency modulation, denoising processing, and closed-loop feedback adjustment, the signal's anti-interference ability and transmission efficiency are significantly improved. In the data analysis process, the present application combines the identification of abnormal values ​​with the refined processing of abnormal data signals, and with the help of DSP technology, it can significantly improve the accuracy and depth of data analysis, and provide a strong guarantee for the accurate identification of abnormal features. This integration of deep analysis and intelligent judgment enables the platform to more accurately evaluate the possible causes and effects of abnormalities, providing a more comprehensive and scientific basis for decision-making. Finally, the present application remotely transmits the abnormal data and its analysis results processed by DSP to the user management platform through the remote communication module, realizing real-time monitoring and management of steel market conditions.

[0079] The above is a further detailed description of the present invention in combination with a specific implementation method, and it cannot be determined that the specific implementation of the present invention is limited to this; for technical personnel in the technical field to which the present invention belongs and related technical fields, based on the technical solution of the present invention, the expansion and replacement of operating methods and data should all fall within the protection scope of the present invention.

Claims

1. A remote analysis and management platform for steel market big data, characterized in that: include: Multi-source data collection module, used to collect raw data related to steel market conditions from multiple channels in real time or at regular intervals; The data extraction and analysis module pre-processes the collected raw data and extracts key information, and uses big data analysis models to identify and analyze outliers in the data; An abnormal data processing module is used to receive the abnormal data signal from the data extraction and analysis module, and transmit the signal to the user management platform after signal conditioning; The storage module is connected to the data extraction and analysis module and is used to store the collected data and analysis results.

2. According to claim 1, a remote analysis and management platform for steel market big data, characterized in that: The abnormal data processing module comprises: A signal conditioning circuit, used for performing complex modulation on the abnormal data signal to optimize signal characteristics; DSP, used to analyze and process the conditioned signal to identify abnormal features in the data; The remote communication module is used to remotely transmit the abnormal data processed by the DSP to the user management platform.

3. According to claim 2, a remote analysis and management platform for steel market big data, characterized in that: The signal conditioning circuit comprises: A signal optimization unit, configured to perform limiting amplification on the abnormal data signal and apply a reference voltage during the amplification process to optimize the dynamic range of the abnormal data signal; A waveform modulation unit, used for converting the low-frequency signal after limited amplification into a high-frequency signal through frequency modulation; A denoising unit, used to remove noise and interference components from the high-frequency signal to improve the signal-to-noise ratio of the signal; The sampling feedback unit is used to sample the denoised signal and implement closed-loop feedback regulation processing on the signal conditioning circuit.

4. According to claim 3, a remote analysis and management platform for steel market big data, characterized in that: The signal optimization unit includes an op amp AR1, an op amp AR2 and an adjustable power supply component. The inverting input terminal of the op amp AR1 is connected to a resistor R1, a resistor R2, a resistor R3, one end of a capacitor C1, a cathode of a diode VD1 and an anode of a diode VD2. The other ends of the resistor R1 and the capacitor C1 are connected to the abnormal data signal input port. The other end of the resistor R2 is grounded. The in-phase input terminal of the op amp AR1 is connected to the adjustable power supply component. The output terminal of the op amp AR1 is connected to the anode of the diode VD1, the cathode of the diode VD2 and the in-phase input terminal of the op amp AR2. The inverting input terminal of the op amp AR2 is connected to the other end of the resistor R3, the output terminal of the op amp AR2 and the denoising unit.

5. According to claim 4, a remote analysis and management platform for steel market big data, characterized in that: The adjustable power supply component includes a voltage zener diode DZ1, an adjustable resistor VR1 and an op amp AR3. The anode of the voltage zener diode DZ1 and one end of the adjustable resistor VR1 are connected to a +12V power supply through a resistor R4, the cathode of the voltage zener diode DZ1 and the other end of the adjustable resistor VR1 are grounded, the adjustment end of the adjustable resistor VR1 is connected to the in-phase input end of the op amp AR3 and one end of the resistor R5, the inverting input end of the op amp AR3 is grounded through a resistor R7, the output end of the op amp AR3 is connected to the other end of the resistor R5, and is connected to the in-phase input end of the op amp AR1 through a resistor R6.

6. A remote analysis and management platform for steel market big data according to claim 5, characterized in that: The waveform modulation unit includes a JFET tube Q1 and a square wave generator. The JFET tube Q1 is arranged between the operational amplifier AR1 and the operational amplifier AR2; the drain of the JFET tube Q1 is connected to the output end of the operational amplifier AR1, the source of the JFET tube Q1 is connected to the in-phase input end of the operational amplifier AR2, and the gate of the JFET tube Q1 is connected to the signal output end of the square wave generator.

7. A remote analysis and management platform for steel market big data according to claim 6, characterized in that: The square wave generator includes an operational amplifier AR4, an inverting input end of the operational amplifier AR4 is connected to one end of a resistor R8 and a resistor R9, and is grounded through a capacitor C2, the other end of the resistor R8 is connected to the anode of a diode VD3, the cathode of the diode VD3 is connected to one end of an adjustable resistor VR2, the other end of the resistor R9 is connected to the cathode of a diode VD4, the anode of the diode VD4 is connected to the other end of the adjustable resistor VR2, a non-inverting input end of the operational amplifier AR4 is connected to one end of a resistor R10 and a resistor R11, the other end of the resistor R10 is grounded, an output end of the operational amplifier AR4 is connected to the other end of the resistor R11, one end of a bidirectional voltage regulator D1, an adjustment end of the adjustable resistor VR2 and a gate of a MOS tube Q1 through a resistor R12, and the other end of the bidirectional voltage regulator D1 is grounded.

8. The remote analysis and management platform for steel market big data according to claim 7, characterized in that: The denoising unit includes a resistor R13, a capacitor C3, a capacitor C4 and an inductor L1, one end of the resistor R13, the capacitor C3 and the inductor L1 is connected to the output end of the operational amplifier AR2, the other end of the resistor R13 and the capacitor C3 is grounded, and the other end of the inductor L1 is connected to the DSP and the sampling feedback unit, and is grounded through the capacitor C4.

9. A remote analysis and management platform for steel market big data according to claim 8, characterized in that: The sampling feedback unit includes a resistor R14 and a resistor R15, one end of the resistor R14 is connected to the output end of the denoising unit, and the other end of the resistor R14 is connected to the inverting input end of the operational amplifier AR3 and is grounded through the resistor R15.

10. The remote analysis and management platform for steel market big data according to claim 1, characterized in that: The data extraction and analysis module uses an isolation forest model to identify outliers in the data.

Images