Mining area water quality multi-parameter testing method and system based on water quality sensor group

Abstract

The invention discloses a mining area water quality multi-parameter test method and system based on a water quality sensor group, and the method comprises the steps: S1, evaluating the zero drift and range drift of the water quality sensor group, and obtaining a zero drift rate alpha, a zero drift amount delta Zi and a range drift confidence interval CI; s2, compensating the zero drift value of the water quality sensor group based on the zero drift amount delta Zi to obtain a compensated zero drift amount Zi '; s3, constructing a multi-modal performance attenuation function D (t) and a Weibull distribution failure probability model Pf (t), and determining the calibration period, the failure standard and the remaining service life of the water quality sensor group according to the multi-modal performance attenuation function D (t) and the Weibull distribution failure probability model Pf (t); s4, monitoring the water quality of the mine in real time, calibrating the water quality sensor group according to the calibration period and the failure standard obtained in the S3, and monitoring the water quality of the mine in real time by adopting the calibrated water quality sensor group to obtain a multi-parameter water quality analysis result; the technical problem of low accuracy of a mining area water quality multi-parameter test analysis method in the prior art is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of water condition and water quality monitoring in mining areas, and relates to a multi-parameter testing method and system for mining area water quality based on a water quality sensor group. Background Art

[0002] With the continuous development of the mining industry, the output of mine water has also increased dramatically. Due to the diversity of my country's geological environment, mine water also presents a complex form. As a special type of industrial wastewater, mine water has a complex and diverse composition, including a large number of constant ions such as calcium, magnesium, sodium, potassium, etc., as well as trace amounts of heavy metal ions such as iron, manganese, lead, cadmium, etc., and a variety of anions such as sulfate, chloride, carbonate, etc. The concentrations of these components in mine water range widely and show significant spatiotemporal variability due to the combined influence of multiple factors such as geological conditions, mining processes, and the surrounding environment of the mine. The indiscriminate discharge of untreated mine water will seriously pollute the ground environment, cause soil compaction and salinization, and pollute surface water sources and ecosystems. The water quality of mine water is directly related to the life safety of miners. Therefore, real-time monitoring and accurate assessment of mine water quality are particularly important. With the expansion of mining scale, the discharge of mine water has increased sharply, and its complex water quality components (such as pH value, heavy metal ions, suspended solids and various anions / cations) pose a serious threat to the ecological environment and miners' safety.

[0003] Among traditional water quality monitoring methods, chemical analysis methods, although highly accurate, require laboratory operations, cannot achieve real-time monitoring, and are time-consuming and labor-intensive (a single test takes 2-4 hours); physical detection methods are susceptible to environmental interference (such as temperature fluctuations causing conductivity measurement errors greater than 10%), and can only obtain a single parameter, making it difficult to comprehensively assess water quality; biological monitoring methods have a slow response speed (the monitoring cycle is as long as 24-48 hours), and biological indicators are easily affected by mine water toxicity, resulting in low reliability.

[0004] In the complex environment of mining areas, the water quality sensor group is exposed to high-mineralization water for a long time, and the zero drift rate of the water quality sensor group can reach 0.5% / day, resulting in the accumulation of measurement errors; due to the interference of multi-parameter coupling, when the temperature rises by 10°C, the pH water quality sensor group reading shifts by 0.1-0.3, and turbidity is nonlinearly related to suspended solids concentration (R 2 <0.8), the traditional fixed-cycle calibration strategy for water quality sensor groups leads to a 30%-50% increase in the frequency of ineffective maintenance and is unable to predict the failure of water quality sensor groups;

[0005] Therefore, in response to the above problems, it is urgent to propose a multi-parameter testing method and system for mine water quality based on a water quality sensor group. By integrating multiple water quality sensor groups, real-time monitoring of multiple parameters in mine water can be achieved. Through dynamic drift model evaluation, segmented dynamic calibration strategy, multi-factor coupling influence matrix construction, Monte Carlo simulation and cost function optimization, etc., the accuracy and reliability of monitoring data can be ensured, maintenance costs can be effectively reduced, and the overall efficiency of the monitoring system can be improved. Summary of the Invention

[0006] In view of the shortcomings of the existing technology, the purpose of the present invention is to provide a multi-parameter testing method and system for mining water quality based on a water quality sensor group, so as to solve the technical problems of low accuracy and efficiency of the multi-parameter testing and analysis methods of mining water quality in the existing technology.

[0007] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0008] A multi-parameter testing method for water quality in a mining area based on a water quality sensor group comprises the following steps:

[0009] S1: Evaluate the zero drift and span drift of the water quality sensor group to obtain the zero drift rate α and zero drift amount ΔZ i and the confidence interval CI of the range drift;

[0010] S2: Based on zero point drift ΔZ i Compensate the zero drift value of the water quality sensor group to obtain the compensated zero drift value Z′ i ;

[0011] S3: Constructing multimodal performance attenuation function D(t) and Weibull distribution failure probability model P f (t), according to the multimodal performance attenuation function D(t) and the Weibull distribution failure probability model P f (t) Determine the calibration period, failure criteria and remaining service life of the water quality sensor group;

[0012] The triggering condition for the multimodal performance attenuation function D(t) is D(t) ≥ K(t)·D_max, where D_max = 0.5 is the maximum allowable attenuation.

[0013] D(t)=w1(ΔZ(t) / Z0)+w2(|S(t)-S_ref| / S_ref)+w3(σ(t) / σ0)

[0014]

[0015] K(t)=0.35+0.15·sin(2πt / 365)

[0016] in:

[0017] S(t) is the upper limit of the range measured by the water quality sensor group at time t;

[0018] S_ref is the range reference value, that is, the standard value of the standard solution;

[0019] σ(t) is the standard deviation of the repeatability error within the sliding time window;

[0020] σ0 is the factory calibration reference value;

[0021] K(t) is the dynamic failure threshold;

[0022] η is the characteristic life;

[0023] β is the shape parameter;

[0024] S4: Monitor the mine water quality in real time. Calibrate the water quality sensor group according to the calibration cycle and failure criteria obtained in S3. Use the calibrated water quality sensor group to monitor the mine water quality in real time and obtain multi-parameter water quality analysis results, including pH value, conductivity, turbidity, and heavy metal ion concentration parameters.

[0025] The present invention also includes the following technical features:

[0026] Step S1 includes the following steps:

[0027] S101: Perform periodic zero point calibration on the water quality sensor group and record the initial zero point drift value Z0 of the water quality sensor group;

[0028] S102: Immerse the water quality sensor group in deionized water and collect n groups of zero point data Z1, Z2, ... Z according to the sampling interval Δt. i …,Z n ;

[0029] Where: ΔZ i =Z i -Z0, ΔZ i represents the absolute drift of the i-th measurement;

[0030] S103: Fit the zero drift curve using the least squares method to obtain the zero drift rate α:

[0031]

[0032] Z i Indicates the zero drift value of the i-th measurement;

[0033] t i Indicates the time between the moment of the i-th measurement and the initial measurement time point t0;

[0034] S104, calculate the zero point drift ΔZ according to the following formula i ;

[0035] ΔZ i =α·t i

[0036] S104: Prepare standard solution in time series t1, t2, ...t i …,t n Collect range data S1, S2, ...S i …,S n ;

[0037] The upper limit of the range of the standard solution is S, the ion concentration of the standard solution is C_s=1.2C_max, and C_max is the range of the water quality sensor group;

[0038] S105: Construct the range drift equation:

[0039] S(t i )=S0·e -λt +ε(t i );

[0040] in:

[0041] λ is the attenuation coefficient;

[0042] ε(t i ) is the random error term;

[0043] S106: Solve the λ value through nonlinear regression and calculate the confidence interval CI of the range drift:

[0044] CI=[λ-k {γ / 2,m-2} SE(λ),λ+k {γ / 2,m-2} SE(λ)]

[0045] in:

[0046] SE(λ) is the standard error;

[0047] k {γ / 2,m-2} is the critical value of the t distribution when the degrees of freedom are m-2 and the significance level is γ.

[0048] Step S2 specifically includes the following steps:

[0049] S201, determine ΔZ i / Z0≤5% is established, if so, the zero offset compensation Zcorrected is calculated using the following formula and the process goes to S203; if not, the process goes to S202;

[0050] Zcorrected=Z i -α·Δti

[0051] in:

[0052] Δt i is the time difference between the moment of the i-th measurement and the moment of the i-1-th measurement;

[0053] S202, determine 5% < ΔZ i / Z0≤15% is established, if so, the zero offset compensation Zcorrected is calculated using the following formula, and the process goes to S203; if not, the water quality sensor group is in an abnormal state and reported to the computer molecular center;

[0054] Zcorrected=Z i -(a·Δt i 2 +b·Δt i );

[0055] in:

[0056] a is the quadratic term coefficient, which characterizes the drift acceleration effect and ranges from [-0.05, 0.02];

[0057] b is the linear coefficient, which represents the drift rate and ranges from [0.8, 1.2];

[0058] S203, using the zero offset compensation Zcorrected to compensate the zero drift Z of the water quality sensor group. i Compensate and get the zero drift Z′ after compensation i .

[0059] In step S3, the calibration period, failure criteria, and remaining service life of the water quality sensor group are determined based on the multimodal performance attenuation function and the Weibull distribution failure probability model, which specifically includes the following steps:

[0060] Q1, the total cost of building a calibration cycle solution C total(t) ;

[0061]

[0062] in:

[0063] Ccal is the direct economic cost of performing a single water quality sensor group calibration operation;

[0064] Cfail is the comprehensive economic loss caused by the failure of the water quality sensor group due to untimely calibration, including direct repair costs and indirect risk costs;

[0065] r is the discount rate;

[0066] L is the design life of the water quality sensor group;

[0067] T is the calibration interval to be optimized;

[0068] Q2, perform N independent Monte Carlo simulations and take the present value of the total cost The calibration period T is calculated using the following formula: opt ;

[0069]

[0070] Q3, determine the failure criteria of the water quality sensor group as HI(t) = 1-D(t) / D_max < 0.45 and N_cal > 50. If HI(t) = 1-D(t) / D_max < 0.45 or N_cal > 50 is satisfied, the water quality sensor group is invalid;

[0071] in:

[0072] HI(t) is the health index;

[0073] N_cal represents the cumulative number of calibrations;

[0074] Q4, the remaining useful life RUL is predicted according to the following formula:

[0075]

[0076] in:

[0077] P acc is the acceptable probability of failure;

[0078] t op The elapsed time.

[0079] Step S2 further includes the following steps:

[0080] S107: Correcting the zero drift rate α using the following formula to obtain a corrected zero drift rate α(T);

[0081] α(M)=α·e -Ea / (R·M)

[0082] in:

[0083] E a is the activation energy;

[0084] R is the gas constant;

[0085] M is the absolute temperature.

[0086] In S3, the values ​​of the weight coefficients w1, w2 and w3 are: w1 = 0.63, w2 = 0.26 and w3 = 0.1.

[0087] A multi-parameter water quality testing system for a mining area based on a water quality sensor group, for implementing the multi-parameter water quality testing method for a mining area based on a water quality sensor group, comprising a water quality sensor group, a signal integrator, a single-chip microcomputer, a computer molecular center, an explosion-proof power supply, and a UPS power supply;

[0088] The output end of the water quality sensor group is connected to the first input end of the signal integrator via an RS485 communication line, the second input end of the signal integrator is connected to the output end of the explosion-proof power supply, the output end of the signal integrator is connected to the first input end of the single-chip microcomputer, the second input end of the single-chip microcomputer is connected to the output end of the UPS power supply, and the output end of the single-chip microcomputer is connected to the input end of the computer molecular center;

[0089] The water quality sensor group is used to monitor the mine water quality in real time, and obtain multi-parameter water quality analysis results, including pH value, conductivity, turbidity and heavy metal ion concentration parameters;

[0090] The signal integrator is used to receive the raw data monitored by each water quality sensor group, convert it into analog signals, and transmit it to the single chip microcomputer;

[0091] The single chip microcomputer is used to calibrate the water quality sensor group and convert the analog signal transmitted by the water quality sensor group into a digital signal and transmit it to the computer molecular center;

[0092] The calibration of the water quality sensor group includes: evaluating the zero drift and span drift of the water quality sensor group, obtaining the zero drift rate α and the zero drift amount ΔZ i And the confidence interval CI of the range drift; used for the zero drift ΔZ i Compensate the zero drift value of the water quality sensor group to obtain the compensated zero drift value Z′ i ; Used to construct multimodal performance attenuation function D(t) and Weibull distribution failure probability model P f (t), according to the multimodal performance attenuation function D(t) and the Weibull distribution failure probability model P f (t) determining the calibration period, failure criteria, and remaining service life of the water quality sensor group; and calibrating the water quality sensor group according to the calibration period and failure criteria;

[0093] The computer molecular center is used to receive and analyze digital signals transmitted by the single chip microcomputer;

[0094] The explosion-proof power supply is used to provide working power for the signal integrator;

[0095] The UPS power supply is used to provide a stable working power supply for the single chip microcomputer and provide a backup power supply when the main power supply fails.

[0096] The water quality sensor group includes a pH water quality sensor, a turbidity water quality sensor, a Ca2+ water quality sensor, a Na+ water quality sensor, a K+ water quality sensor and a Cl- water quality sensor.

[0097] Compared with the prior art, the present invention has the following beneficial technical effects:

[0098] (I) The present invention reduces the long-term measurement error from 8.2% of the traditional method to 1.5% by evaluating and compensating the water quality sensor group before measuring, thereby solving the technical problem of the accuracy of the multi-parameter test and analysis method of mining water quality in the prior art.

[0099] (II) In the present invention, by embedding an embedded Kalman filter in the single-chip microcomputer, the signal-to-noise ratio is improved by 18dB and the dynamic response time is shortened to 200ms, which is suitable for monitoring transient mutations of mine water quality; it helps to ensure the accuracy and reliability of monitoring data and improve the overall efficiency of the monitoring system.

[0100] (III) The present invention constructs a multimodal performance attenuation function and a Weibull distribution failure probability model, determines the calibration period, failure criteria, and remaining service life of the water quality sensor group based on the multimodal performance attenuation function and the Weibull distribution failure probability model, optimizes the calibration interval, and realizes an adaptive calibration period. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] Figure 1 This is a structural diagram of a multi-parameter water quality testing system for mining areas based on a water quality sensor group.

[0102] The specific contents of the present invention are further explained in detail below with reference to the embodiments. DETAILED DESCRIPTION

[0103] It should be noted that, unless otherwise specified, all components in the present invention are components known in the art.

[0104] Specific embodiments of the present invention are given below. It should be noted that the present invention is not limited to the following specific embodiments, and all equivalent modifications made on the basis of the technical solution of this application fall within the protection scope of the present invention.

[0105] The present invention provides a multi-parameter testing method for water quality in a mining area based on a water quality sensor group, comprising the following steps:

[0106] S1: Evaluate the zero drift and span drift of the water quality sensor group to obtain the zero drift rate α and zero drift amount ΔZ i and the confidence interval CI of the range drift;

[0107] S2: Based on zero point drift ΔZ iCompensate the zero drift value of the water quality sensor group to obtain the compensated zero drift value Z′ i ;

[0108] S3: Constructing multimodal performance attenuation function D(t) and Weibull distribution failure probability model P f (t), according to the multimodal performance attenuation function D(t) and the Weibull distribution failure probability model P f (t) Determine the calibration period, failure criteria and remaining service life of the water quality sensor group;

[0109] The triggering condition for the multimodal performance attenuation function D(t) is D(t) ≥ K(t)·D_max, where D_max = 0.5 is the maximum allowable attenuation.

[0110] D(t)=w1(ΔZ(t) / Z0)+w2(|S(t)-S_ref| / S_ref)+w3(σ(t) / σ0)

[0111]

[0112] K(t)=0.35+0.15·sin(2πt / 365)

[0113] in:

[0114] S(t) is the upper limit of the range measured by the water quality sensor group at time t;

[0115] S_ref is the range reference value, that is, the standard value of the standard solution;

[0116] σ(t) is the standard deviation of the repeatability error within the sliding time window;

[0117] σ0 is the factory calibration reference value;

[0118] K(t) is the dynamic failure threshold;

[0119] η is the characteristic life;

[0120] β is the shape parameter;

[0121] S4: Monitor the mine water quality in real time. Calibrate the water quality sensor group according to the calibration cycle and failure criteria obtained in S3. Use the calibrated water quality sensor group to monitor the mine water quality in real time and obtain multi-parameter water quality analysis results, including pH value, conductivity, turbidity, and heavy metal ion concentration parameters.

[0122] In the above technical solution, by evaluating and compensating the water quality sensor group and then measuring it, the long-term measurement error is reduced from 8.2% in the traditional method to 1.5%, which solves the technical problem of the accuracy of the multi-parameter test and analysis method of water quality in mining areas in the existing technology.

[0123] S(t) is the real-time monitoring data of the sensor, obtained through periodic calibration with standard solutions; the value range is determined by the sensor design; S_ref is a laboratory-certified standard solution or international standard; σ(t) is preferably ≤0.5%; σ0 is the calibration data in the factory calibration report; K(t) is the industry standard data, preferably 0.3; η is preferably 12 to 18 months; the preferred value of β: mining sensors usually have β = 1.5 to 2.5 (reflecting progressive wear); in the calibration failure scenario, β = 2.1, which is fitted by experimental data;

[0124] Step S1 includes the following steps:

[0125] S101: Perform periodic zero point calibration on the water quality sensor group and record the initial zero point drift value Z0 of the water quality sensor group;

[0126] S102: Immerse the water quality sensor group in deionized water and collect n groups of zero point data Z1, Z2, ... Z according to the sampling interval Δt. i …,Z n ;

[0127] Where: ΔZ i =Z i -Z0, ΔZ i represents the absolute drift of the i-th measurement;

[0128] S103: Fit the zero drift curve using the least squares method to obtain the zero drift rate α:

[0129]

[0130] Z i Indicates the zero drift value of the i-th measurement;

[0131] t i Indicates the time between the moment of the i-th measurement and the initial measurement time point t0;

[0132] S104, calculate the zero point drift ΔZ according to the following formula i ;

[0133] ΔZ i =α·t i

[0134] S104: Prepare standard solution in time series t1, t2, ...t i …,t n Collect range data S1, S2, ...S i …,S n ;

[0135] The upper limit of the range of the standard solution is S, the ion concentration of the standard solution is C_s=1.2C_max, and C_max is the range of the water quality sensor group;

[0136] S105: Construct the range drift equation:

[0137] S(t i )=S0·e -λt +ε(t i );

[0138] in:

[0139] λ is the attenuation coefficient;

[0140] ε(t i ) is the random error term;

[0141] S106: Solve the λ value through nonlinear regression and calculate the confidence interval CI of the range drift:

[0142] CI=[λ-k {γ / 2,m-2} SE(λ),λ+k {γ / 2,m-2} SE(λ)]

[0143] in:

[0144] SE(λ) is the standard error;

[0145] k {γ / 2,m-2} is the critical value of the t distribution when the degrees of freedom are m-2 and the significance level is γ.

[0146] In the above technical solution, the calibration period T_cal is set to 24±0.5 hours, and zero point calibration is performed under constant temperature (25±0.1°C). The initial zero point reference value Z0 is obtained by using a standard buffer solution, and the output impedance R0 of the water quality sensor group is recorded.

[0147] Preferably, γ is the significance level, usually 0.05, and m-2 is the degree of freedom. In regression analysis or when two parameters are estimated, the degree of freedom is the sample size n minus the number of parameters (here 2).

[0148] The conductivity of deionized water is <0.1μS / cm, and the sampling interval Δt can be 5min.

[0149] λ is the attenuation coefficient, which represents the exponential attenuation rate of the sensor performance, and its value range is 0.001~0.003; ε(t i ) is a standard assumption in statistics, which is a well-known technical

[0150] Step S2 specifically includes the following steps:

[0151] S201, determine ΔZ i / Z0≤5% is established, if so, the zero offset compensation Zcorrected is calculated using the following formula and the process goes to S203; if not, the process goes to S202;

[0152] Zcorrected=Z i -α·Δt i

[0153] in:

[0154] Δt i is the time difference between the moment of the i-th measurement and the moment of the i-1-th measurement;

[0155] S202, determine 5% < ΔZ i / Z0≤15% is established, if so, the zero offset compensation Zcorrected is calculated using the following formula, and the process goes to S203; if not, the water quality sensor group is in an abnormal state and reported to the computer molecular center;

[0156] Zcorrected=Z i -(a·Δt i 2 +b·Δt i );

[0157] in:

[0158] a is the quadratic term coefficient, which characterizes the drift acceleration effect and ranges from [-0.05, 0.02];

[0159] b is the linear coefficient, which represents the drift rate and ranges from [0.8, 1.2];

[0160] S203, using the zero offset compensation Zcorrected to compensate the zero drift Z of the water quality sensor group. i Compensate and get the zero drift Z′ after compensation i .

[0161] In step S3, the calibration period, failure criteria, and remaining service life of the water quality sensor group are determined based on the multimodal performance attenuation function and the Weibull distribution failure probability model, which specifically includes the following steps:

[0162] Q1, the total cost of building a calibration cycle solution C total(t) ;

[0163]

[0164] in:

[0165] Ccal is the direct economic cost of performing a single water quality sensor group calibration operation;

[0166] Cfail is the comprehensive economic loss caused by the failure of the water quality sensor group due to untimely calibration, including direct repair costs and indirect risk costs;

[0167] r is the discount rate;

[0168] L is the design life of the water quality sensor group;

[0169] T is the calibration interval to be optimized;

[0170] Q2, perform N independent Monte Carlo simulations and take the present value of the total cost The calibration period T is calculated using the following formula: opt ;

[0171]

[0172] Q3, determine the failure criteria of the water quality sensor group as HI(t) = 1-D(t) / D_max < 0.45 and N_cal > 50. If HI(t) = 1-D(t) / D_max < 0.45 or N_cal > 50 is satisfied, the water quality sensor group is invalid;

[0173] in:

[0174] HI(t) is the health index;

[0175] N_cal represents the cumulative number of calibrations;

[0176] Q4, the remaining useful life RUL is predicted according to the following formula:

[0177]

[0178] in:

[0179] P acc is the acceptable probability of failure;

[0180] t op The elapsed time.

[0181] In the above technical solution, by constructing a multimodal performance attenuation function and a Weibull distribution failure probability model, the calibration period, failure standard and remaining service life of the water quality sensor group are determined according to the multimodal performance attenuation function and the Weibull distribution failure probability model, the calibration interval is optimized, and an adaptive calibration period is achieved.

[0182] In Q1, Ccal is the direct economic cost of a single calibration, including labor, consumables, and equipment depreciation. Refer to the specific industry standards for the equipment. Cfail is the sum of direct repair costs and indirect risk costs caused by sensor failure. r is the ratio of future costs to present value, reflecting the time value of money. The value range is 5% to 15%. L is the manufacturer's nominal sensor lifespan, generally ranging from 3 to 10 years. T is the time interval between two consecutive calibrations to be optimized.

[0183] In Q2, the i-th simulation process:

[0184] Input parameters: Fixed T (such as T = 30 days), generate random failure time series {t fail,1 ,t fail,2 ,....};

[0185] Cost Calculation fail,n is the specific time point of the nth failure;

[0186] In Q3, if HI < 0.7, a warning is triggered, requiring a shorter calibration cycle; if HI < 0.5, the sensor is deemed failed and needs to be replaced immediately. N_cal is the total number of calibration operations performed during the sensor's lifecycle, which is a known standard and has a maximum threshold of 50.

[0187] In Q4, P acc is the acceptable probability of failure;

[0188] Set the allowable failure probability based on the risk level of the application scenario.

[0189] Value range: High-risk scenarios (mines, chemical industry): ≤5%; general scenarios (municipal water supply): ≤10%; low-risk scenarios (laboratories): ≤20%.

[0190] t op It is the running time, that is, the actual cumulative running time of the sensor, which is recorded by the real-time monitoring system.

[0191] Preferably, the calibration interval T ranges from 6h≤T≤720h.

[0192] Step S2 further includes the following steps:

[0193] S107: Correcting the zero drift rate α using the following formula to obtain a corrected zero drift rate α(T);

[0194] α(M)=α·e -Ea / (R·M) ;

[0195] in:

[0196] E a is the activation energy;

[0197] R is the gas constant;

[0198] M is the absolute temperature.

[0199] In the above technical solution, the influence of temperature on the drift rate α is corrected, thereby improving the accuracy of the dynamic drift model.

[0200] E a is the activation energy, with a typical value range of 20 to 50 kJ / mol; R is the gas constant, the ideal gas constant, which is a fixed constant and matches the unit of activation energy (kJ / (mol·K)); M is the absolute temperature, the temperature of the actual working environment of the sensor;

[0201] In S3, the values ​​of the weight coefficients w1, w2 and w3 are: w1 = 0.63, w2 = 0.26 and w3 = 0.1.

[0202] The present invention provides a multi-parameter water quality testing system for a mining area based on a water quality sensor group, which is used to implement a multi-parameter water quality testing method for a mining area based on a water quality sensor group. The system includes a water quality sensor group, a signal integrator, a single-chip microcomputer, a computer molecular center, an explosion-proof power supply, and a UPS power supply.

[0203] The output end of the water quality sensor group is connected to the first input end of the signal integrator through an RS485 communication line, the second input end of the signal integrator is connected to the output end of the explosion-proof power supply, the output end of the signal integrator is connected to the first input end of the single-chip microcomputer, the second input end of the single-chip microcomputer is connected to the output end of the UPS power supply, and the output end of the single-chip microcomputer is connected to the input end of the computer molecular center;

[0204] The water quality sensor group is used to monitor the mine water quality in real time, and obtain multi-parameter water quality analysis results, including pH value, conductivity, turbidity and heavy metal ion concentration parameters;

[0205] The signal integrator is used to receive the raw data monitored by each water quality sensor group, convert it into analog signals, and transmit it to the single chip microcomputer;

[0206] A single chip microcomputer is used to calibrate the water quality sensor group and convert the analog signal transmitted by the water quality sensor group into a digital signal and transmit it to the computer molecular center;

[0207] Calibration of the water quality sensor group includes: evaluating the zero drift and span drift of the water quality sensor group, obtaining the zero drift rate α and the zero drift amount ΔZ i And the confidence interval CI of the range drift; used for the zero drift ΔZ i Compensate the zero drift value of the water quality sensor group to obtain the compensated zero drift value Z′ i; Used to construct multimodal performance attenuation function D(t) and Weibull distribution failure probability model P f (t), according to the multimodal performance attenuation function D(t) and the Weibull distribution failure probability model P f (t) determining the calibration period, failure criteria, and remaining service life of the water quality sensor group; and calibrating the water quality sensor group according to the calibration period and failure criteria;

[0208] Computer molecular center, used to receive and analyze digital signals transmitted by the single chip microcomputer;

[0209] The explosion-proof power supply is used to provide working power for the signal integrator;

[0210] The UPS power supply is used to provide a stable working power supply for the microcontroller and provide a backup power supply when the main power supply fails.

[0211] In the above technical solution, by building an embedded Kalman filter into the single-chip microcomputer, the signal-to-noise ratio is improved by 18dB and the dynamic response time is shortened to 200ms, which is suitable for transient mutation monitoring of mine water quality. It helps to ensure the accuracy and reliability of monitoring data and improve the overall efficiency of the monitoring system.

[0212] The water quality sensor group includes pH water quality sensor, turbidity water quality sensor, Ca2+ water quality sensor, Na+ water quality sensor, K+ water quality sensor and Cl- water quality sensor.

[0213] Verification example:

[0214] Through single-factor and multi-factor experiments, we determined the key environmental factors that affect the detection accuracy of the water quality sensor group and quantified their impact, providing data support for the multi-factor compensation algorithm.

[0215] (1) Single-factor experiment;

[0216] Control group: The measurement values ​​of the water quality sensor group on the standard solution under ideal conditions (temperature 25°C, pressure 101.3kPa, flow rate 0.5m / s).

[0217] Experimental group: Through single-factor experiments and multi-factor experiments, parameters such as temperature, pressure, and flow rate were changed to analyze their impact on the accuracy of the water quality sensor group.

[0218] Table 1 Single factor experimental results

[0219]

[0220] Conclusion: Temperature and flow rate are single factors that significantly affect the accuracy of the water quality sensor set (p < 0.01), while pressure has a smaller impact (p = 0.03). The error of the water quality sensor set is linearly positively correlated with temperature and flow rate.

[0221] (2) Multifactor experiments;

[0222] Experimental conditions:

[0223] Temperature: 30℃, 40℃

[0224] Flow rate: 1.0m / s, 1.5m / s

[0225] Pressure: Fixed at 101.3kPa

[0226] Table 2 Multi-factor experimental results

[0227]

[0228]

[0229] Conclusion: When multiple factors act together, the error of the water quality sensor group is greater than the sum of the single factors (interaction contribution rate is 14% to 15%). Temperature and flow rate have a synergistic effect, which needs to be corrected by a multi-factor compensation algorithm.

[0230] Table 3 Verification of multi-factor compensation algorithm

[0231] Environmental conditions (T, P, v) Error before compensation (%) Error after compensation (%) Error reduction rate (%) (35°C, 100kPa, 1.2m / s) 2.5 0.7 72％ (45°C, 110kPa, 1.8m / s) 4.1 1.2 70.7％

[0232] Conclusion: The compensation algorithm can reduce the error by more than 70%, verifying the effectiveness of the multi-factor coupling influence matrix. Temperature (weight 0.05) and flow rate (weight 0.1) are the main compensation objects, which is consistent with the single-factor experimental results.

[0233] In summary, temperature (contribution rate 45%) and flow rate (contribution rate 40%) are the core factors affecting the accuracy of the water quality sensor group, and pressure has a smaller impact (contribution rate 15%).

[0234] Interaction: The synergistic effect of temperature and flow rate increases the error by 14% to 15%.

[0235] Compensation effect: The multi-factor compensation model constructed through ridge regression can eliminate 72% of environmental interference errors, and improve the accuracy of the water quality sensor group from ±5% to ±1.5%.

Claims

1. A multi-parameter testing method for water quality in a mining area based on a water quality sensor group, characterized in that: The following steps are involved: S1: Evaluate the zero drift and span drift of the water quality sensor group to obtain the zero drift rate α and zero drift amount ΔZ i and the confidence interval CI of the range drift; S2: Based on zero point drift ΔZ i Compensate the zero drift value of the water quality sensor group to obtain the compensated zero drift value Z′ i ; S3: Constructing multimodal performance attenuation function D(t) and Weibull distribution failure probability model P f (t), according to the multimodal performance attenuation function D(t) and the Weibull distribution failure probability model P f (t) Determine the calibration period, failure criteria and remaining service life of the water quality sensor group; The triggering condition for the multimodal performance attenuation function D(t) is D(t) ≥ K(t)·D_max, where D_max = 0.5 is the maximum allowable attenuation. D(t)=w1(ΔZ(t) / Z0)+w2(|S(t)-S_ref| / S_ref)+w3(σ(t) / σ0) K(t)=0.35+0.15·sin(2πt / 365) in: S(t) is the upper limit of the range measured by the water quality sensor group at time t; S_ref is the range reference value, that is, the standard value of the standard solution; σ(t) is the standard deviation of the repeatability error within the sliding time window; σ0 is the factory calibration reference value; K(t) is the dynamic failure threshold; η is the characteristic life; β is the shape parameter; S4: Monitor the mine water quality in real time. Calibrate the water quality sensor group according to the calibration cycle and failure criteria obtained in S3. Use the calibrated water quality sensor group to monitor the mine water quality in real time and obtain multi-parameter water quality analysis results, including pH value, conductivity, turbidity, and heavy metal ion concentration parameters.

2. The multi-parameter testing method for water quality in a mining area based on a water quality sensor group as claimed in claim 1, characterized in that: Step S1 includes the following steps: S101: Perform periodic zero point calibration on the water quality sensor group and record the initial zero point drift value Z0 of the water quality sensor group; S102: Immerse the water quality sensor group in deionized water and collect n groups of zero point data Z1, Z2, ... Z according to the sampling interval Δt. i …,Z n ; S103: Fit the zero drift curve using the least squares method to obtain the zero drift rate α: Z i Indicates the zero drift value of the i-th measurement; t i Indicates the time between the moment of the i-th measurement and the initial measurement time point t0; S104, calculate the zero point drift ΔZ according to the following formula i ; ΔZ i =α·t i ; S104: Prepare standard solution in time series t1, t2, ...t i …,t n Collect range data S1, S2, ...S i …,S n ; The upper limit of the range of the standard solution is S, the ion concentration of the standard solution is C_s=1.2C_max, and C_max is the range of the water quality sensor group; S105: Construct the range drift equation: S(t i )=S0·e -λt +ε(t i ); in: λ is the attenuation coefficient; ε(t i ) is the random error term; S106: Solve the λ value through nonlinear regression and calculate the confidence interval CI of the range drift: CI=[λ-k {γ / 2,m-2} ·SE(λ),λ+k {γ / 2,m-2} ·SE(λ)] in: SE(λ) is the standard error; k {γ / 2,m-2} is the critical value of the t distribution when the degrees of freedom are m-2 and the significance level is γ.

3. The multi-parameter testing method for water quality in a mining area based on a water quality sensor group as claimed in claim 2, characterized in that: Step S2 specifically includes the following steps: S201, determine ΔZ i / Z0≤5% is established, if so, the zero offset compensation Zcorrected is calculated using the following formula and the process goes to S203; if not, the process goes to S202; Zcorrected=Z i -α·Δt i in: Δt i is the time difference between the moment of the i-th measurement and the moment of the i-1-th measurement; S202, determine 5% < ΔZ i / Z0≤15% is established, if so, the zero offset compensation Zcorrected is calculated using the following formula, and the process goes to S203; if not, the water quality sensor group is in an abnormal state and reported to the computer molecular center; Zcorrected=Z i -(a·Δt i 2 +b·Δt i ); in: a is the quadratic term coefficient, which characterizes the drift acceleration effect and ranges from [-0.05, 0.02]; b is the linear coefficient, which represents the drift rate and ranges from [0.8, 1.2]; S203, using the zero offset compensation Zcorrected to compensate the zero drift Z of the water quality sensor group. i Compensate and get the zero drift Z′ after compensation i .

4. The multi-parameter testing method for water quality in a mining area based on a water quality sensor group as claimed in claim 1, characterized in that: In step S3, the calibration period, failure criteria, and remaining service life of the water quality sensor group are determined based on the multimodal performance attenuation function and the Weibull distribution failure probability model, which specifically includes the following steps: Q1, the total cost of building a calibration cycle solution C total(t) ; in: Ccal is the direct economic cost of performing a single water quality sensor group calibration operation; Cfail is the comprehensive economic loss caused by the failure of the water quality sensor group due to untimely calibration, including direct repair costs and indirect risk costs; r is the discount rate; L is the design life of the water quality sensor group; T is the calibration interval to be optimized; Q2, perform N independent Monte Carlo simulations and take the present value of the total cost The calibration period T is calculated using the following formula: opt ; Q3, determine the failure criteria of the water quality sensor group as HI(t) = 1-D(t) / D_max < 0.45 and N_cal > 50. If HI(t) = 1-D(t) / D_max < 0.45 or N_cal > 50 is satisfied, the water quality sensor group is invalid; in: HI(t) is the health index; N_cal represents the cumulative number of calibrations; Q4, the remaining useful life RUL is predicted according to the following formula: in: P acc is the acceptable probability of failure; t op The elapsed time.

5. The multi-parameter testing method for water quality in a mining area based on a water quality sensor group as claimed in claim 1, characterized in that: Step S2 further includes the following steps: S107: Correcting the zero drift rate α using the following formula to obtain a corrected zero drift rate α(T); α(M)=α·e -Ea / (R·M) in: E a is the activation energy; R is the gas constant; M is the absolute temperature.

6. The multi-parameter testing method for water quality in a mining area based on a water quality sensor group as claimed in claim 1, characterized in that: In S3, the values ​​of the weight coefficients w1, w2 and w3 are: w1 = 0.63, w2 = 0.26 and w3 = 0.

1.

7. A multi-parameter water quality testing system for a mining area based on a water quality sensor group, characterized in that: Used to implement the multi-parameter testing method for water quality in a mining area based on a water quality sensor group as described in any one of claims 1 to 6, comprising a water quality sensor group, a signal integrator, a single-chip microcomputer, a computer molecular center, an explosion-proof power supply and a UPS power supply; The output end of the water quality sensor group is connected to the first input end of the signal integrator via an RS485 communication line, the second input end of the signal integrator is connected to the output end of the explosion-proof power supply, the output end of the signal integrator is connected to the first input end of the single-chip microcomputer, the second input end of the single-chip microcomputer is connected to the output end of the UPS power supply, and the output end of the single-chip microcomputer is connected to the input end of the computer molecular center; The water quality sensor group is used to monitor the mine water quality in real time, and obtain multi-parameter water quality analysis results, including pH value, conductivity, turbidity and heavy metal ion concentration parameters; The signal integrator is used to receive the raw data monitored by each water quality sensor group, convert it into analog signals, and transmit it to the single chip microcomputer; The single chip microcomputer is used to calibrate the water quality sensor group and convert the analog signal transmitted by the water quality sensor group into a digital signal and transmit it to the computer molecular center; The calibration of the water quality sensor group includes: evaluating the zero drift and span drift of the water quality sensor group, obtaining the zero drift rate α and the zero drift amount ΔZ i And the confidence interval CI of the range drift; used for the zero drift ΔZ i Compensate the zero drift value of the water quality sensor group to obtain the compensated zero drift value Z′ i ; Used to construct multimodal performance attenuation function D(t) and Weibull distribution failure probability model P f (t), according to the multimodal performance attenuation function D(t) and the Weibull distribution failure probability model P f (t) determining the calibration period, failure criteria, and remaining service life of the water quality sensor group; and calibrating the water quality sensor group according to the calibration period and failure criteria; The computer molecular center is used to receive and analyze digital signals transmitted by the single chip microcomputer; The explosion-proof power supply is used to provide working power for the signal integrator; The UPS power supply is used to provide a stable working power supply for the single chip microcomputer and provide a backup power supply when the main power supply fails.

8. The multi-parameter water quality testing system for mining areas based on a water quality sensor group as claimed in claim 7, characterized in that: The water quality sensor group includes a pH water quality sensor, a turbidity water quality sensor, a Ca2+ water quality sensor, a Na+ water quality sensor, a K+ water quality sensor and a Cl- water quality sensor.

Images