Water treatment plants and utility companies manage complex operations requiring extensive documentation: water quality reports, regulatory compliance filings, treatment process optimization, customer communications, infrastructure maintenance records, and emergency response plans. Chinese LLMs like DeepSeek V4, GLM-4, and Qwen3 can process this technical data to ensure compliance, optimize processes, and improve customer service. TokenEase's unified API provides water utilities with cost-effective access to these powerful models.
Transform raw water quality data into regulatory-compliant reports with trend analysis and anomaly detection.
import requests
def analyze_water_quality(lab_results, source_water_characteristics, regulatory_limits):
response = requests.post(
"https://tokenease.io/v1/chat/completions",
headers={"Authorization": "Bearer YOUR_TOKENEASE_API_KEY"},
json={
"model": "deepseek-v4",
"messages": [
{"role": "system", "content": "Analyze water quality data for drinking water and wastewater treatment. Identify trends, flag violations, assess treatment effectiveness, and recommend operational adjustments."},
{"role": "user", "content": f"Regulations: {regulatory_limits}\nSource: {source_water_characteristics}\nResults:\n{lab_results}\n\nAnalyze and report."}
]
}
)
return response.json()["choices"][0]["message"]["content"]
results = """
Month: August 2026
Raw water: Turbidity 45 NTU, pH 7.2, TOC 4.5 mg/L, algae count 2,800 cells/mL
Finished water: Turbidity 0.15 NTU, pH 7.8, chlorine residual 1.2 mg/L, TOC 1.8 mg/L
Distribution: Free chlorine 0.8-1.5 mg/L, coliform absent, lead <1 ppb
Trend: Turbidity increasing from 25 NTU in July. Algae bloom detected in source."
source = "Surface water reservoir, agricultural watershed, seasonal algae issues"
limits = "EPA SDWS: Turbidity <0.3 NTU 95% of time. Chlorine residual 0.2-4.0 mg/L. Lead action level 15 ppb."
report = analyze_water_quality(results, source, limits)
Analyze process parameters and chemical dosing data to recommend optimization strategies for coagulation, filtration, and disinfection.
def optimize_treatment_process(current_parameters, influent_conditions, target_effluent, chemical_costs):
response = requests.post(
"https://tokenease.io/v1/chat/completions",
headers={"Authorization": "Bearer YOUR_TOKENEASE_API_KEY"},
json={
"model": "qwen3-235b",
"messages": [
{"role": "system", "content": "Optimize water treatment processes. Recommend chemical dosing adjustments, process changes, and operational improvements to meet effluent targets at lowest cost."},
{"role": "user", "content": f"Costs: {chemical_costs}\nTarget: {target_effluent}\nInfluent: {influent_conditions}\nCurrent:\n{current_parameters}\n\nRecommend optimizations."}
]
}
)
return response.json()["choices"][0]["message"]["content"]
current = """
Coagulation: Alum 8 mg/L, pH adjustment with lime to 7.0
Flocculation: G value 45 s-1, detention 20 min
Sedimentation: Surface loading 0.8 gpm/sf
Filtration: Rate 3.0 gpm/sf, backwash every 48 hours
Disinfection: Free chlorine 2.0 mg/L, CT 120 mg-min/L
"""
influent = "High TOC (4-6 mg/L), moderate turbidity (15-40 NTU), seasonal algae, pH 6.8-7.2"
target = "Turbidity <0.1 NTU, TOC <2.0 mg/L, disinfection byproducts TTHM <80 ug/L, HAA5 <60 ug/L"
costs = "Alum: $0.45/lb. PAC: $1.20/lb. Chlorine: $0.35/lb. Lime: $0.08/lb."
optimization = optimize_treatment_process(current, influent, target, costs)
Generate Consumer Confidence Reports, discharge monitoring reports, and permit compliance documentation.
def generate_compliance_report(report_type, monitoring_data, permit_requirements, facility_profile):
response = requests.post(
"https://tokenease.io/v1/chat/completions",
headers={"Authorization": "Bearer YOUR_TOKENEASE_API_KEY"},
json={
"model": "glm-4-plus",
"messages": [
{"role": "system", "content": f"Generate {report_type} following EPA and state regulatory requirements. Include all required sections, data tables, and public-facing language."},
{"role": "user", "content": f"Facility: {facility_profile}\nPermit: {permit_requirements}\nData:\n{monitoring_data}\n\nGenerate report."}
]
}
)
return response.json()["choices"][0]["message"]["content"]
data = """
2026 Annual Data:
Total coliform: 0 positive samples out of 120 (MCL: <5% positive)
Turbidity: 99.2% of samples <0.3 NTU (MCL: 95%)
Lead: 90th percentile 4.2 ppb (AL: 15 ppb)
Copper: 90th percentile 0.8 mg/L (AL: 1.3 mg/L)
TTHM: Annual avg 58 ug/L (MCL: 80 ug/L)
Chlorine: Range 0.5-2.2 mg/L, avg 1.4 mg/L
"""
facility = "Municipal water system serving 45,000 people, 3 wells + 1 surface water intake, 2 treatment plants"
permit = "Public Water System Permit PWS-12345, NPDES discharge permit NPDES-67890"
report = generate_compliance_report("Consumer Confidence Report", data, permit, facility)
Analyze maintenance logs, inspection reports, and performance data to predict equipment failures and pipe replacement needs.
def predict_infrastructure_needs(asset_inventory, maintenance_history, condition_assessments, performance_data):
response = requests.post(
"https://tokenease.io/v1/chat/completions",
headers={"Authorization": "Bearer YOUR_TOKENEASE_API_KEY"},
json={
"model": "kimi-k2",
"messages": [
{"role": "system", "content": "Predict water infrastructure maintenance needs. Analyze pipe age, material, break history, and water quality impacts to prioritize replacements and repairs."},
{"role": "user", "content": f"Performance: {performance_data}\nConditions: {condition_assessments}\nHistory: {maintenance_history}\nAssets:\n{asset_inventory}\n\nPredict needs."}
]
}
)
return response.json()["choices"][0]["message"]["content"]
assets = """
Zone A: 12 miles cast iron, installed 1950s-60s, 8 breaks past 5 years
Zone B: 8 miles ductile iron, installed 1980s, 2 breaks past 5 years
Zone C: 15 miles PVC, installed 2000s, 0 breaks
Zone D: 5 miles galvanized steel, installed 1940s, 15 breaks past 5 years, lead service lines present
"""
history = "Zone D: frequent discolored water complaints, pressure complaints. Cathodic protection failing."
conditions = "Zone A: internal tuberculation reducing capacity 20%. Zone D: external corrosion severe, wall thickness 40% of original."
performance = "System water loss: 18% (target <12%). Zone D accounts for 35% of total losses."
prediction = predict_infrastructure_needs(assets, history, conditions, performance)
Draft clear, empathetic customer communications about service disruptions, rate changes, and water quality issues.
def draft_customer_communication(topic, affected_customers, technical_details, communication_tone):
response = requests.post(
"https://tokenease.io/v1/chat/completions",
headers={"Authorization": "Bearer YOUR_TOKENEASE_API_KEY"},
json={
"model": "deepseek-v4",
"messages": [
{"role": "system", "content": f"Draft utility customer communications that are {communication_tone}. Explain technical issues clearly without jargon. Include actionable information and timelines."},
{"role": "user", "content": f"Tone: {communication_tone}\nCustomers: {affected_customers}\nTopic: {topic}\nTechnical:\n{technical_details}\n\nDraft communication."}
]
}
)
return response.json()["choices"][0]["message"]["content"]
topic = "Boil water notice due to temporary treatment plant malfunction"
affected = "2,500 households in northern service area, primarily residential, includes 2 schools and 1 nursing home"
technical = "Coagulation basin mechanical failure caused turbidity to spike to 2.5 NTU for 45 minutes. System automatically diverted to backup basin. Finished water now meets standards but precautionary notice required per protocol."
tone = "calm, informative, empathetic, authoritative"
communication = draft_customer_communication(topic, affected, technical, tone)
Review and update emergency response plans based on new threats, regulatory changes, and lessons from incidents.
def analyze_emergency_plan(current_plan, recent_incidents, new_threats, regulatory_updates):
response = requests.post(
"https://tokenease.io/v1/chat/completions",
headers={"Authorization": "Bearer YOUR_TOKENEASE_API_KEY"},
json={
"model": "qwen3-235b",
"messages": [
{"role": "system", "content": "Analyze water utility emergency response plans. Identify gaps, recommend updates, and ensure compliance with AWWA standards and EPA requirements."},
{"role": "user", "content": f"Regulations: {regulatory_updates}\nThreats: {new_threats}\nIncidents: {recent_incidents}\nPlan:\n{current_plan}\n\nAnalyze and recommend updates."}
]
}
)
return response.json()["choices"][0]["message"]["content"]
plan = "ERP last updated 2023. Covers: contamination events, natural disasters, cyber threats, power outages. Exercises conducted annually."
incidents = "2025: Cryptosporidium detection in source required 48-hour boil notice. Communications delays identified."
threats = "Increased cyber threats to SCADA systems. Climate change causing more frequent extreme weather. Emerging contaminants (PFAS) detected regionally."
regs = "EPA requiring enhanced cybersecurity assessments. New PFAS MCLs finalized. Bioterrorism preparedness updates."
updates = analyze_emergency_plan(plan, incidents, threats, regs)
Get $1 free credits (1M tokens) to automate water quality reporting and optimize treatment processes.
Start with TokenEase →